Refugee Impacts on Turkana Hosts A Social Impact Analysis for Kakuma Town and Refugee Camp Turkana County, Kenya Varalakshmi Vemuru, Rahul Oka, Rieti Gengo, and Lee Gettler Refugee Impacts on Turkana Hosts A Social Impact Analysis for Kakuma Town and Refugee Camp Turkana County, Kenya Varalakshmi Vemuru, Rahul Oka, Rieti Gengo, and Lee Gettler NOVEMBER 2016 Copyright © 2016 The International Bank for Reconstruction and Development/ The World Bank Group 1818 H Street, NW Washington, DC 20433, USA All rights reserved. First printing: November 2016 Manufactured in the United States of America. Cover photo: Entrepreneurial women; Photo credits (cover and text): Rahul Oka, University of Notre Dame The findings, interpretations, and conclusions expressed in this report are entirely those of the authors and should not be attributed in any manner to the World Bank, or its affili- ated organizations, or to members of its board of executive directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. Contents Acknowledgments.................................................................................................................... vii Abbreviations and Acronyms .................................................................................................... ix 1. Introduction: The Need for Deeper Ethnographic Analysis ................................................... 1 2. Goals, Objectives, and Methodology.................................................................................... 5 Methodology ............................................................................................................................. 5 Ethnographic Research .............................................................................................................. 6 Psychosocial Stress and Anthropometric Surveys in the Turkana Host Community.................. 7 Expectations and Hypotheses.................................................................................................. 11 Caveats and Study Limitations ................................................................................................. 13 3. Historical Context of Turkana County and Kakuma Refugee Camp ...................................... 15 Precolonial, Colonial, and Postcolonial Transitions: From Rulers to Ruled, from Pacified to Excluded ........................................................................................................ 15 Kakuma Refugee Camp, 1992–2015 ........................................................................................ 21 4. Interactions at Kakuma ...................................................................................................... 27 Narratives ................................................................................................................................ 29 Interactions and Narratives ..................................................................................................... 40 External Factors and Narratives of Suffering, Resilience, and Dignity ..................................... 43 Perceptions as Outcomes ........................................................................................................ 44 5. Measuring Impacts ............................................................................................................ 47 Measuring Nutritional and Psychosocial Well-Being Among the Turkana ............................... 47 Body Mass Index ..................................................................................................................... 49 Sum of Skinfolds ...................................................................................................................... 51 Psychosocial Stress .................................................................................................................. 56 6. Gender-Differentiated Norms, Dynamics, and Constraints at Kakuma ................................ 63 Gendered Vulnerability and Structural Violence among Turkana Men of Kakuma ................. 65 Gender-Based Violence and Structural Violence among the Turkana Women of Kakuma ..... 67 iii iv Contents 7. Devolution, Oil, and Water: Perspectives and Narratives on Underdevelopment, Recent Discoveries, and Structural Changes ....................................................................... 73 Narratives and Responses to Devolution, Oil, and Water ....................................................... 74 Feedback between Emergent and Previous Narratives ........................................................... 76 Narrative Summary ................................................................................................................. 77 8. Modeling the Pathways of Interactions and the Complex Impacts of Refugees on Host Communities ........................................................................................................ 79 Narrative-Interaction Model for Refugee–Host Relationship .................................................. 81 Bringing It Together ................................................................................................................. 82 Finding 1: History, Experience, Narratives, and Interactions Shape One Other....................... 83 Finding 2: Complex Interactions .............................................................................................. 83 Finding 3: Refugees Have A Net Positive Impact on Turkana Hosts at Kakuma ....................... 84 Finding 4: Variations Underlying Positive and Negative Impacts ............................................ 85 Finding 5: Turkana Attitudes, Evaluations, and Hopes ............................................................ 85 9. Conclusion ......................................................................................................................... 87 References ............................................................................................................................... 91 Appendixes .............................................................................................................................. 95 A Mean and Variation of Body Mass Index and Sum of Skinfolds ........................................... 97 B Regression of Body Fat Percentage for Women and Men ................................................. 105 C Worry Codes for Principal Component Analysis ................................................................ 109 D Regression of Sum of Skinfolds Tests ................................................................................. 111 E Background Paper: The Future of Kakuma’s Youth—Opportunities and Challenges ......... 115 F Background Paper: Gender Dynamics in Turkana County and Kakuma Refugee Camp .................................................................................................................... 145 G Background Paper: Implications of a Changing Social Order and Structures for Peace and Development .............................................................................................. 169 H Background Paper: Opportunities for Sustainable Livelihoods........................................... 199 I Background Paper: Socioeconomic Mapping and Analysis of Kakuma............................... 229 Boxes H.1 The Myth of Atanayeche .................................................................................................... 201 H.2 Building a Herd................................................................................................................... 204 H.3 The Rise and Fall of Herd Size............................................................................................. 209 H.4 Profile of a Small Shop Owner............................................................................................ 211 H.5 Profile of a Business Woman.............................................................................................. 212 Contents v Figures 3.1 Population of Kakuma Refugee Camp (1993–2016)............................................................. 25 4.1 Perceptions of Interactions with Refugees Among the Turkana by Age and Gender .......... 41 4.2 Proportion of Host Community Reporting Negative (Nonviolent) Refugee Impacts............ 44 4.3 Proportion of Host Community Members Reporting Refugee Violence or Mistreatment........................................................................................................................ 45 4.4 Proportion of Host Community Members Reporting Positive Refugee Impacts.................. 46 5.1a Average SSF for Men and Women across Turkana County................................................... 52 5.1b Average SSF for Young, Middle-aged, and Older Women Across Turkana County............... 52 5.1c Average SSF for Young, Middle-aged, and Older Men across Turkana County..................... 52 5.2 Average BFP Across Sites...................................................................................................... 55 5.3 Worries Among the Host Community in Kakuma, Disaggregated by Gender and Age......... 57 5.4 Density and Variation of Worries Across Locations.............................................................. 58 5.5 Principal Components Analysis of Worries Across Sites....................................................... 59 8.1 Trends in Positive and Negative Perceptions of Refugees among the Turkana of Kakuma, Lorengo, Lokichoggio, Lodwar, and Lorugum......................................................... 80 8.2 Narrative-Interaction Model................................................................................................. 82 E.1 Total Primary School Enrollment (Class 1–8) by Gender in Kakuma, 2000–08................... 125 G.1 Perceptions Regarding the Economic Impacts of Refugees ............................................... 178 G.2 Perceptions of Refugee Social Impact................................................................................. 179 G.3 Concerns of Hosts Across Sampling Locations.................................................................... 181 G.4 Plot of Principal Components 1 and 2 ............................................................................... 182 G.5 Plot of Principle Components 2 and 3................................................................................ 183 H.1 Total Businesses ................................................................................................................. 217 H.2 Sum of Skinfolds among the Turkana (Mean Value, Age-Controlled)................................. 223 H.3 Psychosocial Stressors Experienced by Turkana in Kakuma, Loki, Lorengo and Lorugum....................................................................................................................... 225 I.1 Number of Businesses Per Type of Services/Goods........................................................... 236 I.2 Tukana County Budget Breakdown, 2015/16..................................................................... 248 Maps 2.1 Map of Turkana County, Kenya............................................................................................... 7 3.1 Kakuma Refugee Camp, Layout Map.................................................................................... 22 7.1 Oil (Red) and Water (Blue) Discoveries in Turkana County................................................... 75 vi Contents Tables 2.1 Research Areas and Key Questions for the Social Impact Analysis......................................... 8 3.1 Major Famines and Droughts in Turkana.............................................................................. 20 5.1 BMI Across Four Locations.................................................................................................... 49 5.2a Average BMI of Population in Turkana Compared with the World Health Organization’s Threshold for Malnourishment/Undernourishment..................................... 50 5.2b Average BMI by Age and Gender Group............................................................................... 50 5.3a Average BMI by Age Group................................................................................................... 51 5.3b Average BMI by Age and Gender Group............................................................................... 51 5.4a Average SSF Measurements and Variations in SSF by Location and Gender........................ 53 5.4b Results of T-Tests Showing Difference in SSF Across Sites.................................................... 53 A.1 Average Body Mass Index and Sum of Skinfolds for Women............................................... 97 A.2 Average Body Mass Index and Sum of Skinfolds for Men..................................................... 99 E.1 Drought Magnitude in Lodwar, Turkana (1950–2008)........................................................ 118 E.2 Tree-Crown and Shrub Cover by Distance from the Kakuma Refugee Camp...................... 119 E.3 Livestock Population in Turkana District (1993–2009)........................................................ 120 E.4 Education by County, Constituency, and Ward in Turkana West (percentage)................... 126 E.5 Type of Work Done by Level of Education.......................................................................... 129 E.6 Employment and Gender Dynamics of the Employment Initiatives................................... 138 E.7 Perceptions about Causes of Violence in Turkana.............................................................. 139 Acknowledgments This report was written by a World Bank team led by Varalakshmi Vemuru in col- laboration with a consultant team from University of Notre Dame comprising Rahul Oka, Rieti Gengo and Lee Gettler. The study team was led by Varalakshmi Vemuru, Senior Social Development Specialist, World Bank, and the field research, data analysis, and report preparation components were led by Dr. Rahul Oka with assistance from Rieti Gengo and analytical support from Dr. Lee Gettler and Dr. Mark Golitko, all from the University of Notre Dame. The accompanying background papers were written by Teresia Katindi Njonjo (appendix E), Verena Phipps-Ebeler (appendix F), Margarita Puerto Gomez and Samhita Kumar (appendix G), Varalakshmi Vemuru (appendix H), and Camille Hennion (appendix I). This study is part of the joint World Bank–United Nations High Commissioner for Refugees (UNHCR)-led report entitled “Yes” In My Backyard: The Economics of Refu- gees and their Social Dynamics in Kakuma, Kenya, and builds on this social impact analysis (Sanghi et al. 2016). For overall guidance and support, we are grateful to Dr. Apurva Sanghi (Task Team Leader, World Bank), Raouf Mazou (Country Representative, UNHCR Kenya), John Wagacha Burton (Technical Coordinator, UNHCR Kenya), and Yonatan Araya (Senior Regional Policy Officer, UNHCR). Angelina Musera provided excellent logistical and administrative assistance in Nairobi. We thank the UNHCR Kakuma Sub-Office, especially Inge de Langhe, Stanley Thuku, Julia Githiomi, Peter Ekal, Peter Echukut, Peter Lorukia, and Winston Otieno for invaluable logistical and coordination support in the field while conducting this two- part research. We thank the Kellogg Institute for International Studies and the Institute for Scholar- ship in the Liberal Arts at the University of Notre Dame for funding the Psychosocial Impact Assessment component of the Social Impact Analysis. vii viii Acknowledgments We would also like to thank our helpers, translators, and research assistants in the field who enabled us to make the deep connections and enduring relationships that allowed access to accurate and precise data. Specifically, among the refugee community, we’d like to thank Kamal Nur Isack, Said Kuli, Edin Mohammed, Mohammed Guyo, Ahmed Abdi, Safia Hassan, Suad Hussain, Ahmed Dahir, Mesfin Getahun, Menelik Hawasa, and Halima Ahmed. Among the host community, we thank Ewoton Arumah, Moses Lomuria, Anne Onim, Ekamais Bruno, and Jackson Nakhusa for their patience in teaching us the intricacies and complexities of Turkana culture and for keeping us safe in difficult areas. Finally, we thank the numerous people we interviewed in Kakuma Refugee Camp and Turkana County who invited us into their homes and, regardless of their situation, gen- erously shared their knowledge, experiences, attitudes, hopes, and lives, as well as their food. Abbreviations and Acronyms BMI body mass index KDHS Kenya Demographic and Health Survey km kilometers mm millimeters NGO nongovernmental organization PCA principal component analysis SSF sum of skinfolds UN United Nations UNHCR United Nations High Commissioner for Refugees WFP World Food Programme ix Introduction: 1 The Need for Deeper Ethnographic Analysis Mass displacements of people due to natural disasters, such as floods, droughts, earth- quakes, volcanoes, hurricanes, and typhoons, or anthropogenic disasters, such as fam- ines, wars, conflicts, persecution, or genocide, are common throughout history (Jones and Murphy 2009; Sanghi et al. 2010). Various historical records from ancient to recent societies indicate that displaced peoples have always sought access to nearby or neigh- boring areas for safety and security and for basic humanitarian needs, including food, water, shelter, and clothing. In response to these migrations, destination societies have historically recognized their duty to host displaced peoples until they can be repatriated, resettled, and/or integrated (Oliver-Smith 1996). Grievances are especially Often, the host society’s views on the refugees are driven by fear prevalent in protracted situations and uncertainty of “outsiders,” “migrants,” or “foreigners,” feeding a of refugee camps built in rural dominant narrative that refugees and displaced peoples usually have areas with virtually no local negative impacts on their host communities (Agier 2002, 2011; Veney infrastructure. 2007). There is a crucial need to understand the complexities of these interactions with respect to the dominant narratives and the frequent occurrence of violence and conflict over real and perceived grievances between refugees and their local host communities with regard to access to land, water, jobs, opportuni- ties, and development (Betts 2009; Jacobsen 2002; Martin 2005; Porter et al. 2008). Grievances are especially prevalent in protracted situations of refugee camps built in rural areas with virtually no local infrastructure (Agier 2002, 2011; Crisp 2003). In these circumstances, refugees find themselves surrounded by impoverished local host com- munities underserved by regional and national governments and with little to no role in making decisions about the land and resources given to the refugees by the host govern- ments and the UNHCR. Host communities subsequently view the refugee camps as an imposition (Aukot 2003; Chkam 2016; Jacobsen 2002), and these grievance narratives have existed as long as the refugee camps have. The role that daily interactions between refugees and hosts play are often ignored in such narratives, including those between refugee and host, employer and employee, trader and consumer, or patron and client— as well as master and servant or oppressor and victim (Betts 2014; Grabska 2011). The 1 2 Refugee Impacts on Turkana Hosts: A Social Impact Analysis resulting complexities result in emergent experiences of Indignity and servitude as well as dignity and well-being (Aukot 2003; Hilhorst and Jansen 2009). This report outlines the complex dynamic interface between narratives, interactions, and external factors in determining the impacts of refugees on the social economies and psychosocial well-being of host communities by focusing on the refugees of Kakuma Refugee Camp and the Turkana host community located in Turkana County, Kenya. Various factors are explored in the attitudes, behaviors, and actions of refugee-host interactions that foster a sense of indignity (matharau in KiSwahili and ng’imeny in Turkana). The feeling of servitude and inequity many experience as a result of these interactions can be described by the Turkana word etic (pronounced e-tich), which means working for or under. Turkana women with refugee neighbors 1. Introduction 3 Turkana County, until 2013 known formerly as Turkana District of the Rift Valley Prov- ince of Kenya, is the largest county in Kenya (77,000 square kilometers). It is also one of the most impoverished and marginalized areas of the country and indeed in all of Sub- Saharan Africa. Kakuma Refugee Camp is the largest settlement in Turkana County. Due to the crises in Somalia (2011–14) and South Sudan (2013–ongoing), the camp has grown from a population of 85,000 in 2011 to more than 160,000 in September 2016. The complex interactions between the Turkana The complex interactions host community and the refugees are being tested by the changing between the Turkana host demographics of the refugee settlement at Kakuma with regard to its community and the refugees scale and the cultural origins of the refugees. In addition to coping are being tested by the changing with an influx of more than 100,000 people in the past five years, the demographics of the refugee Turkana community must also deal with recent changes in governance settlement at Kakuma. infrastructure that are expected to bring significant development to Turkana County, including devolution, resource discovery, and oil extraction in South Turkana County, and fresh water aquifers in North Turkana County under the follow- ing assumptions: n The devolution process and the increased autonomy of the county government with regard to resource allocation and spending will help build and support public devel- opment infrastructure at an unprecedented rate, scale, and reach; n The discovery of oil will lead to the development of a regulated extractive industry that provides opportunities, jobs, and a better standard of living for the local people; and n The discovery of immense freshwater aquifers will enable the development of exten- sive and intensive agriculture, including irrigation-based methods, potentially lead- ing to long-term development. The combined effects of these three recent developments could potentially transform Turkana County’s public infrastructure and social and political economies from their current state of general impoverishment, marginalization, lack of health care and nutri- tion, and low levels of education. 4 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Acknowledging that, at least for the short to medium term, Kakuma Refugee Camp may be an enduring part of the Turkana landscape, with the ongoing presence of more than 160,000 refugees possessing diverse skill sets and varying access to capital and resources, allows for the possibility that refugees could play a large and positive role in the projected development of Turkana County. To create an inclusive space for refu- gees and hosts to work together for the generation of sustainable livelihoods and lives, policy makers need to understand and accommodate the complex interactions between host and refugee. Specifically, the interactions must move from generating indignity (matharau or ng’imeny) to fostering dignity (heshima in KiSwahili and arimatoi in Tur- kana); and for the Turkana, moving from servitude (etic) to working with or together (aking’arakin). These complexities are mapped and described in this social impact analysis based on intensive ethnographic and survey research conducted by the World Bank, UNHCR, and the University of Notre Dame. 2 Goals, Objectives, and Methodology This social impact analysis describes the complexities of the interactions between refu- gees and their host community and assesses their positive and negative outcomes within the current relief paradigm, contextualized by: (1) the history of interactions between the Turkana people and the central Kenyan government from the British colonial period to the current administration; (2) recent developments regarding devolution, oil, and water; and (3) since 1992, the arrival and continuing flow of large numbers of refugees into northern Turkana. An analysis of the data helps to map the strengths and weak- nesses with ongoing interactions, and examines potential problems and intervention points. The overall goal of the social impact analysis is to provide a set of evidence-based insights and policy recommendations to the Kenyan government, the UNHCR and asso- ciated bodies, and development actors to support Turkana’s development and to benefit Kakuma so that members of refugee and host communities alike can study, find jobs and opportunities, run businesses, work, raise children, celebrate achievements, and mourn their dead according to their own traditions and customs, and that they have equal access to health care and other public services. In short, the larger goal is to empower the refugees of Kakuma as well as their Turkana hosts to live lives of dignity. Methodology According to the World Development Report (World Bank 2015: 144–45) “ethnography [is a] powerful tool for understanding the ways in which social and cultural context shapes decision making, choices, and interpersonal relations,” and for a greater under- standing of motivations, perceptions, agendas, and subsequent actions and responses to any development intervention. Using an ethnographic approach to gain data on social and economic behaviors, motivations, perceptions, and agendas of both the host and refugee communities, the primary objective of the social impact analysis is to assess and analyze the net social—including psychosocial and socioeconomic—costs and benefits of the presence of refugees at the Kakuma Refugee Camp on the host community of Turkana County. The findings from the social impact analysis in conjunction with the economic impact analysis will inform and enable development and relief experts, policy makers, and social scientists to better understand the evolution of refugee–host 5 6 Refugee Impacts on Turkana Hosts: A Social Impact Analysis interactions over the past 24 years, and it will assist UNHCR, the Kenyan government, and other interested organizations initiate and execute the proposed transformation of Kakuma through the necessary shift from a short-term emergency approach to a long- term approach to development and financing. Ethnographic Research To better understand the social economies of the Turkana people and the refugees of Kakuma, we used ethnographic approaches, following a longitudinal study conducted in and around Kakuma since 2008, composed of participant observations, semistructured interviews, and focus groups with the Turkana in their homes and villages, including following a traditional moran cattle-raiding near the Lokichoggio area, which did not result in any casualties. In addition to interactions during the interviews for this study, participant observations involved close engagement with the host and refugee commu- nities in food-sharing and other friendship rituals before, during, and after the interac- tions rather than conducting one-time interviews, which are more common in these kinds of studies. This research benefitted from the resulting deep trust built between study participants and researchers, which has: (1) enabled a deeper understanding of Turkana cultural institutions, such as bond-friendships and Turkana “cattle complex” (Herskovits 1926), which are “foreign” even to non-Turkana Kenyans; (2) uncovered attitudes among the Turkana regarding their 100 years of exclusion and marginaliza- tion/discrimination; and (3) elicited accurate and precise data on the complexity of the attitudes, experiences, and narratives of the Turkana regarding the refugees. The ethnographic research was conducted in the towns of Kakuma, Lodwar, Lorugum, and Lokichoggio in Turkana County, Kenya (map 2.1) in May and June 2015. Kakuma town is adjacent to the refugee camp and is the primary residence of the members of the host community who work in the refugee camp as construction workers, domestic ser- vants, security guards, charcoal and firewood producers and sellers, livestock producers and sellers, and even survival sex workers.1 Lorugum, a town 170 kilometers (km) south of Kakuma, is an area of sustained development and significant investment by the Kenyan government due to the Turkwel River Hydroelectric Project. Development efforts include investments in reforestation, livestock management, agriculture, and small industry. Lokichoggio is 96 km from Kakuma and 35 km from the Kenya–South Sudan Border. Lokichoggio, the site of the UNHCR refugee camp from 1990–92, also served as the pri- mary base for the UNHCR Sudan Relief Mission and, subsequently, for numerous non- governmental organizations (NGOs). After 2008, UNHCR and the NGOs involved in the 1. Survival sex is not considered a financial transaction. It involves exchanging sex for basic subsistence needs. 2. Goals, Objectives, and Methodology 7 South Sudan relief and humanitarian mission moved to Juba, and the Kenyan work staff moved to Kakuma or back to southern Kenya, leaving Lokichoggio an underdeveloped and impoverished community. Lodwar is the capital of Turkana County and the seat of its government. It has grown in political importance as a result of the governance changes brought about by devolution. As the county’s capital, Lodwar has also served as the center for development and disaster and conflict relief efforts for the Turkana people, and it hosts many relief and development organizations at a level comparable to Kakuma. The research followed procedures for informed consent and voluntary participation and protection of informants as described by the Institutional Review Board of the Office of Research at the University of Notre Dame (Protocol ID: 15–05–2534). The field research was sanctioned by the Kenyan government and UNHCR. Data were elicited using par- ticipant observations, intensive semistructured interviews, and focus group discussions. The participants (n=121) interviewed were selected from the refugee (n=30) and the host communities (n=91), primarily through contacts from previous research in Kakuma (see Ohta 2005; Oka 2014; Horn 2010). Precautions were taken to make sure that initial participants and assistants were selected from different ethnic and social groups within both the refugee and host communities. While these initial contacts and research assis- tants might have heard about each other, care was taken that they lived and worked in social networks that did not overlap, Map 2.1. Map of Turkana County, Kenya which also ensured that the participants would hail from dif- ferent social networks. Seven themes were selected for the eth- nographic research under the social impact analysis (table 2.1). These themes were built around previous ethnographic work on the informal and formal economies of Kakuma and Turkana County (Pérouse de Montclos and Kagwanja 2000; Ohta 2001, 2005; Oka 2011, 2014). Psychosocial Stress and Anthropometric Surveys in the Turkana Host Community A large-scale survey of stress and health indicators was con- ducted through interviews and anthropometric measure- ments in Kakuma town, which is adjacent to Kakuma Refugee Camp, as well as in the towns of Lokichoggio, Lorengo, and Lorogum in Turkana County, Kenya (see map 2.1). The meth- odologies for the health and nutritional survey of Turkana were built on previous bioanthropological approaches in South Turkana County (Brainard 1986; Pike 1999, 2004). Source: HRW 2015. 8 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Table 2.1. Research Areas and Key Questions for the Social Impact Analysis Area of Research Research Questions 1. Socioeconomic n What is the demographic information of the host and refugee communities, disaggregated mapping by gender, age, ethnicity, and nationality? n What are the intrahousehold livelihood roles and contributions as well as assets among the refugee and host communities? n What are the existing means of livelihood and access to capital, skills training, and markets among the host and refugee communities? n How have the livelihood means evolved for the host community since the independence of Kenya? How have the arrival and protracted stay of refugees impacted this evolution? n What type of access to natural resources do the host and refugee communities have in order to meet their household consumption and livelihood needs? n What are the positive and adverse or negative coping strategies, such as child labor or survival sex, that have been adopted by members of both the host and refugee communities? In what ways have the arrival and protracted stay of refugees impacted coping strategies among the host community? 2. Social services n What is the education and health status among the host and refugee communities? n What is the current level of access to health and education services among the host and refugee communities? n What is the status of drinking water and sanitation facilities among the host and refugee communities? 3. Social organization n What is the social composition of the host and refugee communities? and structure n What are the characteristics of vulnerable households? n What is the status and level of interaction among social groups differentiated by gender, ethnicity, age, religion, and nationality, within and between the host and refugee communities? n What are the formal and informal leadership structures in the host and refugee communities? n How have the social organization and structures in the host community evolved since the independence of Kenya? In what ways have the arrival and protracted stay of refugees impacted this evolution? (continued) 2. Goals, Objectives, and Methodology 9 Table 2.1. Continued Area of Research Research Questions 4. Economic n What is the level of economic participation among host and refugee community participation households? What formal and informal livelihood activities do refugee and host communities pursue? n What are the constraints to economic participation? n What are the political economy implications of policy changes to address these constraints? n What are the levels of access to and confidence in systems governing economic participation, such as grievance redress and arbitration, among host and refugee communities? n What is the status of input and output markets for locally produced goods, including agricultural and natural resource-based products? n What is the status of connectivity for host and refugee communities—locally, regionally, nationally, and internationally? n What are the key means of transportation? Are they affordable? 5. Community n What recent changes have been implemented in the prevalent rules, incentives, and social organizations and norms that govern interactions, including the adverse impacts of corruption, rent seeking, institutions and bribes? n What are the roles and functions of community organizations? n What is the level of access to benefits from formal and informal institutions among the host and refugee communities? 6. Conflict and violence n What is the overall security situation for the host community, given the location of Turkana County? n What types and how prevalent is conflict and violence among the host and refugee communities, including sexual and gender-based violence? What are their causes and drivers? n Who are the perpetrators and who are the survivors of violent activities? n What conflict resolution and peace-building mechanisms and institutions currently exist? How effective are they? 7. Development n Who are the major developmental actors in the host and refugee communities? activities n How are planning and implementation responsibilities organized between them to avoid duplication and ensure synergy and efficiency? n What are the perceptions of the host and refugee communities regarding the activities and benefits received due to developmental interventions aimed at both? n How have the interventions of development actors changed in response to the situation on the ground? n What are the expectations of the host and refugee communities regarding development interventions? 10 Refugee Impacts on Turkana Hosts: A Social Impact Analysis A robust sample (n=600) was assembled comprising equal numbers of men and women from each of the four towns. A frequent assumption is that refugee camps negatively impact their host communities through resource competition and conflict. Based on prior anthropological research, this research focused on whether or not host commu- nities could also benefit from the presence of refugees through economic exchange. To assess the impact, anthropometric measures of physical and nutritional well-being between communities in the Turkana were compared. Data were elicited for the follow- ing aspects: n Demographic information, including gender, age, marital status, number of spouses, number of children, livestock, livelihood, and social support. n Psychosocial stress, for which participants were asked to freely list the main sources of stress and worry in their lives. Specifically, following a previous study among Tur- kana (Pike 2004; Pike and Williams 2006), each informant was asked to list their worries and concerns about their lives with no limit on their number and without ranking them. The host community was also asked about refugee-related worries and perceptions across the four sites in order to calculate negative, positive, and neutral perceptions toward the refugees. While the informants came up with their own names and descriptions for their worries, the strength of classification/categori- zation of them into the eight categories of stressors was tested using Cohen’s Kappa for pair-wise comparisons of agreement (Cohen’s K = 0.88, p<0.0001). n Anthropometric measures, including body mass index (BMI; kg/m2) based on height and weight and sum of skinfolds (SSF). While BMI provides a convenient proxy for physical well-being and energy status, SSF, which measures body fat con- tent, is a well-established indicator of energy status. Body fat content, especially in impoverished and malnourished populations, is positively correlated with better nutrition, physical health, and maternal and reproductive health. There are various ways to measure body fat content, including isotopic determination of total body water, radiography, electrical conductance, and impedance. However, the most com- mon methods are hydrostatic weighing, which involves whole body immersion, and SSF, which is less accurate or precise than hydrostatic measurements but which is ideal for field studies because it is a standardized nonintrusive method of measuring the thickness of subcutaneous fat tissue at 2–7 points on the body. It enables the calculation of body fat percentage (BFP), which allows for a determination of energy status for the prior 6–12 months. 2. Goals, Objectives, and Methodology 11 The specific aim was to test whether Kakuma Refugee Camp affects the well-being of the Turkana host community, measured through BMI, SSF, and BFP (n=600). The methodology for this part of the social impact analysis involved detailed surveys of 300 men and 300 women, comprising 75 men and 75 women from each of the four study locations. Expectations and Hypotheses The primary hypotheses tested using the data from the social impact analysis are grounded in observations from previous studies, which conclude that the impact of the presence of a large number of refugees in Kakuma has both positive and negative impacts. Specifically, the study expected to find that: n There is a significant variation in the well-being of the host community at Kakuma compared with those of the other study sites—Lorugum, Lorengo, Lokichoggio, and Lodwar. The hypothesis was tested with the following caveats: – There is some overlap among Kakuma, Lodwar, and Lorugum due to the close presence of relief and development efforts and organizations; and – Lorengo and Lokichoggio have the worst outcomes in terms of social, physical, and mental well-being due to the absence or withdrawal of development and relief organizations, respectively. n Pathways taken by the host community of Kakuma toward well-being differ from the host communities at the other study sites, confounded by the impact of the refugee presence and by daily interactions with refugees, hosts, and relief-development orga- nizations. Specifically: – The presence of the relief-development infrastructure and the refugees provide the Turkana of Kakuma greater access to jobs, wage labor, and other opportu- nities, but increase the potential for conflict and distrust between refugees and hosts. – Despite increased access to jobs, better wages, and development opportunities in Kakuma, there is still a significant variation in physical and psychosocial well- being within the host community, indicating exacerbated structural inequity. 12 Refugee Impacts on Turkana Hosts: A Social Impact Analysis n Levels of development investment would emerge as a significant factor in under- standing the differential trajectories of well-being across Turkana County, where the level of development investment and presence is directly correlated with social, economic, psychosocial, and physical well-being. n There would be widespread distrust in the ability of the government and associated bodies or organizations to deliver the promised benefits of devolution and the oil and water discoveries in Turkana. n There would be a general disenfranchisement of youth and women in Turkana and a growing distance between political and social elites and Turkana’s other residents. n Structural and physical violence would be disproportionately manifested along the categories of gender and age, whereby women and youth would be more negatively affected by the systemic exclusion from participating in social, economic, and politi- cal spheres. Shopping in Mogadishu, Kakuma 2. Goals, Objectives, and Methodology 13 The results of the social impact analysis allowed for the investigation of these expecta- tions and the expansion of the analysis to include previously unconsidered variables, including the role of dignity/indignity, which seems to significantly affect interactions between refugees and their hosts. Caveats and Study Limitations Due to the sensitive nature of the questions and volatility of the research area, neither the ethnographic nor the psychosocial surveys used random sampling methods. Both data collection approaches utilized the researchers’ contacts with trusted participants known for 2–7 years to engage new participants. Within-group bias was mitigated by the fact that the prior contacts were drawn from a variety of ethnic, religious, social, political, and economic groups, and responses were crosschecked within and between groups. The data were collected in a semistructured and focus group interview format based on specific predetermined questions. Interviews lasted at least two hours, and many participants were interviewed repeatedly. Researchers engaged with participants through food-sharing and friendship rituals before, during, and after the interactions, ensuring room for variation and further discussion of their responses, including clarifi- cations, verifications, explanations, and descriptions with specific examples. The psychosocial stress and physical health assessments used a questionnaire-based format in addition to anthropometric measurements to elicit the maximum amount of data, followed by focus group interviews to contextualize the data with each group. We are aware that by generally contributing to undernourishment and malnutrition, vari- ous factors, including water, sanitation, and hygiene, effect nutritional status and hence anthropometric measurements. This is due to chronic or recurring diarrhea and, impor- tantly, a more recent condition known as tropical or environmental enteropathy, which is significantly correlated with exposure to “large quantities of fecal bacteria” (Mahmud and Mbuya 2016: 20; see also Humphrey 2009). The Turkana at all four sites live in these types of environments because they practice ablutions in water holes and rivers frequented by livestock and domestic animals. In Kakuma, the local Turkana forage for water from the riverbed and share it with livestock and domestic animals, and with a population density of more than 2,000 persons/km2, it is quite likely that water, sanita- tion, and hygiene impacts nutritional health. However, because all four sites are in areas with a high potential for fecal contamination—with Kakuma potentially being the most contaminated—we would expect that the lack of water, sanitation, and hygiene would dampen any nutritional benefits offered by the camp to the Turkana host community. In the next chapter, we contextualize the refugee–host social economies and inter- actions in the history of Turkana, especially over the last century. Historical Context of 3 Turkana County and Kakuma Refugee Camp Turkana County is regarded as one of the cradles of human evolution, especially with the recent discoveries of stone tools dating to about 3.3 million years ago—the old- est in the world (Harmand et al. 2015). The region also boasts of some of the finest technological development in subsistence, trade, and production in East Africa for the past 12,000 years (Wright et al. 2015). History suggests that, prior to 1900, the ances- tors of the current Turkana were the lords of their realm, with control over territory that included Turkana County and large areas of bordering Uganda, South Sudan, and Ethiopia, as well as areas within Kenya to the south and the east of the county (Broch- Due and Hodgson 2000; Lamphear 1988). Nevertheless, the current state of social and economic well-being in the county suggests that the people of Turkana are among the most impoverished and marginalized communities in Sub-Saharan Africa and the world (Anderson and Broch-Due 2000). The bewildering disparity between the well- being of the Turkana in 1900 and their impoverishment in 2015 begs the question: How did the conditions of the Turkana decline so dramatically? Precolonial, Colonial, and Postcolonial Transitions: From Rulers to Ruled, from Pacified to Excluded Archaeological and oral history studies suggest that from 1500 to 1890 CE, the Tur- kana pastoralists came from the more fertile areas of southeastern Uganda, specifically the Karamoja Escarpment, which expands across the western basin of Lake Turkana. Between 1895 and 1910 CE, the British colonial administration moved along with the railroad linking the coast to Western Kenya, reaching Mt. Elgon by the late 1880s (Lam- phear 1988; Wright et al. 2015). After the conquest of the Mt. Elgon region in 1895 CE, the British began to expand northward toward Turkana, presumably to link their west- ward expansion in East Africa to their southward expansion in the Sudan. The ferocity of the Turkana people was the political justification used for military pacification to “save” the Turkana from themselves and from growing Ethiopian imperialism, as well as to protect the Luhya, Sabaot, Pokot, Luo, Samburu, and Rendille pastoralists from Turkana raids (Lamphear 1992: 23). 15 16 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Initially, the Turkana proved a real match for the British (1900–05), which led to a change in policies involving the punitive confiscation of livestock in order to “mitigate” its raiding. The most significant and crippling outcome of the British actions was that, even as the Turkana were forced into more arid areas west of Lake Turkana, they lost tremendous amounts of wealth due to the depletion of their livestock (Mburu 2001). Significant outcomes of the British–Turkana interactions include: n Changes in subsistence activity. Changes in subsistence patterns were driven by livestock depletion and, ironically, the intensification of pastoralism as a primary subsistence activity that conferred status, as well as a decline in agriculture and trade. – Livestock depletion. By 1900, the rinderpest2 epidemic across East Africa had significantly depleted Turkana livestock levels. The confiscation of livestock over the next two decades further depleted the herd sizes to levels below the asset threshold.3 This left the herd vulnerable to further attrition and depletion due to disease and catastrophes, particularly recurring droughts and famines. – Intensification of pastoralism. The depletion of livestock was correlated with greater investment in pastoralism as the primary and often the only subsistence activity deemed appropriate among the Turkana. Specifically, the ability to build, maintain, and maximize herd sizes became the determinants of status among Turkana men. – Decline in agriculture. Various areas in Turkana County, including the Lokichoggio area, the Tarach River, and the Turkwel River have long been centers of agricultural production for the Turkana. Agricultural activities are primarily conducted by the women who grow crops such as sorghum as well as forage for wild foods. According to the foundational Nayece myth of the Turkana as well as informants for this study, these processed and stored foods used to be the main sources of nutrition among the Turkana. The intensification of pastoralism and its associated status led to a decline in the status of women’s activities—such as agriculture. 2. Rinderpest is an acute infectious disease among cattle caused by a morbillivirus among cattle. 3. The asset threshold is a narrow range for calculating optimal herd size. It varies based on environment and political economy (Murphy 2014). If the herd size is above the asset threshold, then the rate of attrition among livestock through raiding, death, and/or disease is lower than the replacement rate. Herd sizes below the asset threshold usually indicates that the attrition rate is higher than the replacement rate. 3. Historical Context of Turkana County and Kakuma Refugee Camp 17 – Decline in trade. The Turkana were involved in trade networks at local and regional levels across the western Lake Turkana basin and where livestock, grain, weapons, tools, ornaments, clothing, and other goods were exchanged. The isola- tion and the militarization and/or supervision of the Turkana area in the colo- nial and postcolonial periods eroded these networks, leaving the Turkana overly dependent on external aid, on development initiatives, and on pastoralism. As stated by Wright et al. (2015: 357): “In the past, interdependent social networks evolved over vast areas to mitigate risk associated with the changing climate. However, as the local economy [became] increasingly dependent on outside resources [after the British incursions], there [were and] are fewer incentives to cooperate within the neighborhood and more incentives to fight a zero- sum game for primacy in controlling access to those resources.” n Changes in social organization. Changes include the alteration in community structure and gender relationships. – Community and household structure. The emergence of the awi-adakar sys- tem, which continues to characterize the current Turkana settlement patterns, emerged in the colonial period. n Awi. An awi is an extended family headed by an older man who controls the awi’s herd with his brothers, sons, and nephews, with the (usually older) men practicing polygamy and maintaining multiple wives. n Adakar. An adakar comprises five or six awis belonging to one patrilineage that gather as a larger group to graze and water their animals at previously chosen sites. These annual gatherings also served as zones for communal ritu- als, marital alliance brokering, rites of passage ceremonies, and other cultural activities. n Alomar/arumrum. These are small hamlets consisting of 15–20 individual awi households or two to three adakars. As Turkana groups were forcibly settled, the residential patterns were disjointed with women living, raising children, and tending young animals in rural semipermanent hamlets of individual awis and men migrating around 200–500 kilometers (km) per year to graze their animals. Lorengo one of the sites in the psychosocial study, is an example of this. 18 Refugee Impacts on Turkana Hosts: A Social Impact Analysis – Exchange relationships. n The creation of kinship exchange systems means that any progress made by an individual Turkana has to be shared across the larger extended kin group, including but not limited to help with money, food, jobs, and other benefits within the power of the individual family member. n Bond-friendships, a holdover from pastoral systems, involves a small “ask” that serves as the foundation for enduring relationships but which may be interpreted by outsiders as “begging” (Gulliver 1951, 1955; Ohta 2005). – Gender relationships. A number of factors significantly altered the sociopolitical landscape of Turkana, such as forced resettlement into small villages, transhu- mance (affecting men in particular), and the emergence of politically appointed headmen and chiefs, which led to a decline in many Turkana cultural institutions, including the ngimoruk (diviners), and then further led to erosion of Turkana cultural identity (Fratkin 2001). n Growth of a warrior culture. The depletion of average herd size along with the status-driven intensification of pastoralism led to an increase in livestock- raiding by groups of young warrior men (moran) and other militia activities (ngoroko). These served as efficient ways to quickly rebuild depleted herds and establish warrior status. Contrary to most depictions of East African pastoral groups, livestock raiding became a normative institution among the Turkana after the colonial encounter, as did the incipient violence between the Turkana and their neighbors. n Determination of status among men. As previously noted, pastoralism became one of few socially sanctioned activities that allowed men to gain sta- tus and recognition in their society. For younger men, status was determined by their achievements in raiding activities and in contributing to the awi’s abil- ity to pay bridewealth. 3. Historical Context of Turkana County and Kakuma Refugee Camp 19 n Decline in the status of women. The forcible settlement of the Turkana also relegated women to gendered activities that offered no attached status, such as childbearing and rearing, raising of young livestock, practicing agriculture, and foraging. The primary concern for many young women to this day appears to be collecting enough beads in their neck jewelry to entitle them to a high dowry. One informant from Kakuma recounted the popular local understand- ing of the lower roles given to the women after the colonial period: “When the British fought the Turkana and they would round up everyone, they would lock up the women and children together in houses and send the men to work, to do hard labor. Women were just supposed to be at home and take care of kids. Maybe this made men come up with a psychology that women cannot do things and are like kids.” (KII 2015) Ethnographic studies suggest that by the 1950s, there was a clear difference among even neighboring groups with regard to type and specialization of subsistence practices (Fratkin 1986, 1997; Gulliver 1951, 1955). The forcible settlement of the Turkana destabilized pastoral movement pathways and forced the ethnic The forcible settlement of the groups of the Western Lake Turkana Basin into small disjointed residen- Turkana also relegated women to tial hamlets occupied by specialized subsistence groups lacking regional gendered activities that offered exchange networks and increasingly dependent on externally obtained no attached status, such as resources over which they had no control. childbearing and rearing, raising of young livestock, practicing In the late 1950s, the British colonial authorities “forcibly moved the agriculture, and foraging. southernmost elements of the Turkana back to Turkana District presum- ably to ensure the ethnic homogeneity of ‘traditional’ Samburu domain,” but in reality, the aim was to send the Turkana back to the closed district that was excluded from development and that kept them from participating in anticolonial movements. Super- vision policies ranged from requiring passes to leave the Turkana district to mandatory registration of anyone traveling with the police (Simpson 1996: 692). This gave rise to the expression “naenda Kenya,” which literally means “I’m going to Kenya,” but which really encapsulates the life of residents excluded from a nation that demands their loy- alty while refusing them the rights and privileges of citizenship. The Turkana continued to be categorized as a closed district, and the Turkana were barred from freedom of movement until 1986, more than two decades after Kenya achieved its independence. 20 Refugee Impacts on Turkana Hosts: A Social Impact Analysis The northern part of Kenya, including Turkana County, has suffered from increasingly severe droughts lasting for years, resulting in the depletion of grazing lands and water resources (Bush 1995). Combined with the lack of public infrastructure and services, droughts grow into famines with high mortality rates among humans and livestock. Because the ability of pastoralists such as the Turkana to manage their resources is closely linked to prevailing environmental conditions, notable famine events are often given names as they become part of the cultural folklore. Most of the informants in this study were very familiar with the names of the famines, the associated stories of suf- fering and resilience, and the impacts on their own families and herds. Major famines and droughts in Turkana that experienced high rates of livestock mortality (measured as percentage of deaths—see table 3.1) shows that between 1950 and 2008, there were 12 droughts, five of which were moderate to extreme. The 2008 drought persisted until 2011, growing into a vast famine that stretched from Turkana to Mandera, Wajir, Mars- abit, and Garissa, and by June–July 2011, into Somalia. The region has experienced two additional years of consecutive drought since then, in 2013 and 2014. These drought and famine events have recurred regularly for multiple years with small intervals of around 5–6 years separating them. The narrowness of the intervening periods without drought means that maintaining or replenishing livestock numbers after a drought or Table 3.1. Major Famines and Droughts in Turkana Mortality Drought Rate among Magnitudea Livestock Year (1950–2008) Name of Drought Name of Famine (percent) 1950 -0.43 1952 -2.84 Lotiira (1952–57) Lotiira 61 1960 -0.65 Namotor Namotor 55 1964 -0.18 1971 -1.09 Kimududu Kimududu/Kibekbek 54 1976 -0.27 Kibekbek 54 1980 -6.25 Kiyoto atang’aa (1980–84) Kiyoto atang’aa/Lopiar 65 (1980–87) 1986 -0.28 Lopiar (1986–87) 53 1990 -6.0 Lokwakoyo/Akalkal (1990–95) Lokwakoyo/Akalkal 53 1999 -3.31 Logara/Epompo (1999–2001) Logara/Epompo (2000) 63 2005 -0.06 2008 -0.59 Source: “Background Paper: The Future of Kakuma’s Youth—Opportunities and Challenges” (appendix E of this report). a. Drought magnitude is the cumulative sum of the consecutive negative standard precipitation index values for a number of years. Standard precipitation n index values of -2.0 and less are extreme. Values of -1.0 to -1.5 indicate moderate drought. 3. Historical Context of Turkana County and Kakuma Refugee Camp 21 famine is incredibly difficult. The six named droughts and famines show a high correla- tion between livestock mortality and catastrophic malnutrition, forced displacement, and human mortality. Kakuma Refugee Camp, 1992–2015 The Kakuma area was formerly a watering and meeting hole for Turkana pastoralists from surrounding areas. The name Kakuma means “the place of the giant tortoise.” (the Turkana word for tortoise is akumurae). It was a seasonal communal meeting ground for various adakars of the Turkana to water and graze their livestock during the wet season. Prior to 1992, Kakuma was a culturally and economically significant location for the Turkana pastoralists living within 200 km of the town; it grew into a small town (population: 5,887) with a livestock market controlled primarily by the Somali traders. It served as a rest-and-fuel stop for truck drivers on the A1 highway that links Kitale in Western Kenya to Juba, which is now in South Sudan. Subsequently, Kakuma town had restaurants, hotels, garages, petrol stations, and general stores. The large shops and petrol stations were owned and operated by Somali traders, while the smaller shops, garages, restaurants, and hotels were owned by Turkana and Meru traders. The relationship between the few outsiders at Kakuma and the Turkana residents prior to 1991 was built on over three decades of mutually beneficial interactions, particularly with the Somali trading firms established in Kakuma since the 1960s (Oka 2014; Oka et al. Under review). The firms operated as middlemen for two types of exchanges at the livestock markets in Kakuma—between the rural producers and the Kakuma market and between the Kakuma market and the ones in Lodwar, Kitale, and Eldoret. They established wholesale firms—still active in Kakuma today—to supply other local and regional business and customers with dry groceries and hardware (Oka et al. Under review). Oral historical data suggest that the Somalis would “load up donkeys with sugar, tumbaku [tobacco], beans, leather/rubber slip- pers, metal plates [for arrowheads and spears], and clothing, and go into the bush. When they came back, they would be leading many livestock that they had traded for.” (KII 2010) The Somalis were known for their honesty and for their love of the Turkana, making them “blood brothers who would many times go raiding with the young men, for livestock.” Kakuma Refugee Camp was formally established in 1992 on the A1 highway on the northwestern bank of the Tarach River (map 3.1). The camp was built to accommodate people fleeing the conflict in Sudan that lasted from 1985–2005 (Pérouse de Montclos 22 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Map 3.1. Kakuma Refugee Camp, Layout Map 3. Historical Context of Turkana County and Kakuma Refugee Camp 23 and Kagwanja 2000). Lokichoggio was the site of the initial UNHCR refugee welcom- ing station and until 2005, the base of the United Nations (UN) South Sudan relief. However, it was only 35 km away from the conflict zone and cross-border raids from the Jie of Uganda and the Toposa of South Sudan along the Kenya–Sudan border—too close for adequate refugee protection and services. After engaging in consultations with the Kenyan government as well as local leaders and elders of the Turkana com- munity, UNHCR chose Kakuma, 96 km to the southeast of Lokichoggio, as its new center of operations. By the end of 1992 and 1993, the first group of refugees—the Nuer—reached Kakuma, followed in 1993 and 1994 by the Dinka, Ethiopian Amhara, Ethiopian Oromo, and some Somalis. This led to increasing panic among Kakuma area residents, stemming from two main points of contention: 1. The Turkana were concerned that the people who were fighting each other would bring their conflict to Kakuma, particularly the Sudanese Nuer and Dinka. Accord- ing to an informant, recounting the events in 1992: “In Sudan, people are killing each other. We will not agree to letting dangerous strangers come and settle nearby” (Ohta 2005: 229). 2. The interactions of UNHCR and the Kenyan government with the local people mainly consisted of meetings with government appointed chiefs, so the voice of the local people was not heard. One informant, recalled: “I was 12 years old when the camp was built, but I remember when the refugees first came here, our people reacted and were angry, because they were going to occupy our lands. What will happen to our livestock? I remember them asking our elders.” (KII 2015) Another woman said: “When the camp displaced the Turkana, no compensation was given to the people who were displaced. If it was given, then it went to government of Kenya, not to us. Mostly, it’s because the Turkana do not have title deeds. Among Turkana, the land belongs to the community, but people’s identity is tied to the land. We always know the first residents, we could call them the owners.” (KII 2015) The Turkana were also upset about having to give up their annual meeting place on the eastern and western sides of the Tarach River, which intersected with Kakuma town and the A1 highway. The place had economic and political importance as a seasonal meeting 24 Refugee Impacts on Turkana Hosts: A Social Impact Analysis site for adakar gatherings as well as mythic significance (Lamphear 1988). However, along with fear and anger, there was also conciliation, mercy, and anticipation of new opportunities. One informant who said that people got angry when the refugees were coming, added in the same interview: “The government approached our leaders and asked our leaders, who said, ‘Let’s welcome the refugees, let them stay, they’re suffering because of their war. We’ll host them.’” (KII 2015) Anthropologist Itaru Ohta reports that just before the camp was established, “Several young men brought other news […] from Kakuma: they said that many people would be employed in the construction work for the refugee camp.” One of them, who had previously been employed doing road construction work told me excitedly, “We can get a big job. We will be paid 50 shillings a day.” He asked [Ohta] for some scissors to cut his hair, which was woven in the traditional style, in order to get a job. He was wearing a shirt and trousers that day, although he usually wore only a sheet around his body (Ohta 2005: 229). Within a year after the establishment of the Kakuma Refugee Camp, which in 1993 comprised the areas now known as Kakuma Camp 1 (referred to as Mogadishu, Addis, and Hong Kong), numerous Ethiopian Amhara, Oromo, Somali, and some Sudanese Nuer and Dinka refugees had set up retail shops, restaurants, and barber shops, offering refugees and Turkana alike access to goods at retail prices. The goods were mainly imported by three Somali firms in Kakuma town, who had by then established reliable supply lines with suppliers in Kitale, Eldoret, Nakuru, Nairobi, and Mombasa. Originally built for around 80,000 Sudanese refugees, the population of Kakuma Refu- gee Camp has fluctuated from 35,000 at its establishment to 80,000 in 2009 and over 1,600,000 in 2016 (figure 3.1). The camp has taken in people fleeing conflicts from Burundi, Congo, Rwanda, Ethiopia, Eritrea, and Uganda. In 1997, after the destruction and closing of the Utanga-Benadiri camp, Kakuma received a large Originally built for around inflow of Somali refugees, including Somali Barawa, Bajuni, and Bantu. 80,000 Sudanese refugees, the population of Kakuma Refugee After the Sudan Peace Accords in 2005, it was hoped that the bulk of Camp has fluctuated from 35,000 the South Sudanese refugees in Kakuma would be safely repatriated. at its establishment to 80,000 in Indeed, between 2006 and 2008, more than 40,000 Sudanese refugees 2009 and over 1,600,000 in 2016. were repatriated to various areas in the present day South Sudanese territories, including the towns of Bor, Kapoeta, Narus, Bahr-el- Ghazal, and Juba (Grawert 2010). The population of Kakuma suddenly dropped to 45,000, depressing local businesses and leaving behind abandoned and broken home- steads. However, as shown in figure 3.1, the population of Kakuma rebounded due to: 3. Historical Context of Turkana County and Kakuma Refugee Camp 25 Figure 3.1. Population of Kakuma Refugee Camp (1993–2016) 200 Population (in thousands) 150 100 50 0 1990 1995 2000 2005 2010 2015 Year Source: Data from UNHCR (2015) and Ohta (2002, 2005). (1) violence by Al Shabaab and the famine in Somalia (2011–13), when 45,000 Somali refugees pushed the population from 85,000 to 130,000; and (2) the continuing violence and lack of infrastructure in South Sudan between 2010 and 2011, which resulted in a reverse flow of refugees back into Kenya and Kakuma (Oka 2014). In December 2013, South Sudan erupted into violence, resulting in a full-scale con- flict with a huge death toll and major displacement. The utter destabilization of South Sudan’s reconstruction has displaced over a million people, scattered at internally dis- placed persons (IDP) camps in South Sudan and at refugee camps in Uganda, Ethiopia, Sudan, and Kenya, particularly Kakuma. Since January 2014, Kakuma has received almost 80,000 South Sudanese refugees. Furthermore, the official counts at Kakuma have systematically underestimated the actual population because of the concealed populations of unregistered migrants living in the refugee camp seeking access to the relief, medical, nutritional, and educational services. Estimates of this unofficial latent subpopulation vary from 5,000–15,000, with at least one undocumented person per family currently residing in the camp. Between January and December 2014, almost 50,000 new South Sudanese refugees flee- ing the ongoing violence were registered in Kakuma and given shelter, mainly in Kakuma 3 and 4 (new areas further to the north and west built in 2009 and 2010 to house the incoming Somali refugees). The capacity of the various civic, relief, and development organizations are being taxed to their limits, necessitating changes in the processes by which relief and development services are managed and administered by the Kenyan government, UNHCR, and other organizations. The increased refugee population in Kakuma combined with the growing population of Kakuma town (urban population of 26 Refugee Impacts on Turkana Hosts: A Social Impact Analysis 31,932 in 2009) also means that the active interactions and encounters between the ref- ugees and the Turkana hosts can only intensify—positively and negatively. This means that any proposed changes to relief and development policies in Kakuma that affect refugees and the Turkana must account for and accommodate the different experiential and narrative outcomes of these interactions, especially the negative ones that could lead to violence and conflict, as they have in the past. The next chapter explores and analyzes various categories of interaction, drawing on ethnographic and survey data collected between 2008 and 2015. The remains of repatriation 4 Interactions at Kakuma The arrival of 35,000 refugees (1992–93) transformed the socioeconomic dynamics of Kakuma town. By 2000, the town’s population had almost doubled to an estimated 9,000, as had the camp’s—to an estimated 80,000 due to the influx of 20,000 refugees from Sudan, including Darfur; 15,000 from Somalia due to the closing of Utanga Bena- diri Camp in Mombasa; and the rest from the Democratic Republic of Congo, Eritrea, Burundi, Rwanda, and Uganda. Even in 1998, when the official count was 58,000, the population density of the occupied areas of Kakuma 1 and Kakuma 2 was roughly 400 persons/hectare (Montclos and Kagwanja 2000; Ohta 2005). Along with the increased avenues for consumption of goods and services and growing cosmopolitanism, the refugee presence also fundamentally altered the livestock herd- ing and consumption system among the Turkana. The flood of cereal-based food aid through Oxfam’s North Turkana Drought Relief Programme in Kakuma, Lokichog- gio, and Kalokol (1992–94) led to a decrease in the pressure for slaughtering livestock. This was paralleled by the high demand for meat among the (relatively) cash-rich refugees passing through Lokichoggio to Kakuma who were Along with the increased receiving money through remittances transmitted by hawala banks avenues for consumption of and agents. (Practiced predominantly in the Middle East and Horn of goods and services and growing Africa, this system of value transfer is based on large networks of money cosmopolitanism, the refugee brokers.) According to Bush (1995: 257), the price of meat rose from K presence also fundamentally Sh 6/kg ($0.06 USD) in 1992 to K Sh 9/kg in 1993 and K Sh 22/kg in altered the livestock herding and 1994. This might have led to the development of an entirely new group of consumption system among Turkana brokers, the nimuchurus. The nimuchurus operate at two levels the Turkana. with respect to the livestock marketing associations: (1) external: those who travel to smaller markets or wait on the outskirts of towns seeking sellers with animals; and (2) Internal: those who operate within the livestock marketing associa- tion market yards. The external nimuchurus sell to the internal nimuchurus who then trade among themselves until they reach a particular price ceiling, at which point the animal/s are offered to the external buyers. The external buyers, mainly Somalis along with some Ethiopians, Congolese, and Burundians, then negotiate the price with the 27 28 Refugee Impacts on Turkana Hosts: A Social Impact Analysis final nimuchurus. This system has left the small producers and herders entirely at the mercy of the brokers, buyers, butchers, and consumers. Some refugees began providing small jobs to the Turkana, including domestic labor, such as sweeping, dusting, washing dishes and clothes, fetching water, and process- ing food; casual and manual labor, such as construction of houses and fences, main- tenance of buildings and roads, and portage; and service jobs, such as restaurant and shop cleaners and sweepers. The UNHCR and other relief and development organiza- tions generated jobs in a number of areas, including security, such as watchmen and police; construction; maintenance; transport; clerical; and, occasionally, management. In addition to jobs provided directly by the refugees of the relief-development complex, the people of Turkana found opportunities in supplying basic necessities to the camp, including firewood and charcoal, fencing and housing materials (thornbushes and tree poles), and occasionally livestock in the form of trade with individual refugees. In addition to these economic interactions, the Turkana engage with the refugees in friendship and through marital alliances, especially with the South Sudanese Dinka and Nuer. As various informants noted, the Turkana acknowledge similarities between themselves and the South Sudanese, including that they are both pastoralism-centered cultures, that livestock management plays a primary role in terms of being a subsistence activity, and that they have a mutual understanding of the role of conflict and aggression in daily interac- tions. Engagement with other groups such as the Congolese, Rwandans, Burundians, Somalis, and Ethiopians usually consist of the Turkana as consumers or workers in the shops and homes of refugees. There are numerous examples of enduring friendships between individual fami- lies and people, usually formed and reinforced through regular meetings in public arenas, such as churches, bars, restaurants, cafes, teahouses, and sporting fields. Ohta (2001, 2005) sug- gests that the narratives of conflict and violence between the two groups serve to obfuscate the fact that between 1992 and 2005, the Turkana and the refugees established complex, mutually Weddings and marriages: life goes on 4. Interactions at Kakuma 29 beneficial, and enduring relationships. This study extends Ohta’s analysis by document- ing positive and negative interactions between the Turkana and the refugees as reported or observed by Turkana residents of Kakuma (n=150). The data extends beyond Ohta’s research to suggest multiple outcomes of the complex interplay between experiences and narratives that range from cooperation to conflict, and where individual interac- tions are shaped by the larger narrative but also by the history of individual-level inter- actions. The range of interactions between the Turkana and refugees is broad and varies from positive to negative in terms of outcomes and perceptions that create narratives, which in turn shape interactions. Data from multiple sources suggest that narratives of conflict and cooperation are not mutually exclusive. Rather, they bleed into each other, forming a metanarrative in which the refugees and the Turkana alike simultaneously perceive each other through lenses of compassion, friendliness, resentment, and enmity, but largely under the narrative of the violent “other.” An analysis of the data sheds light on the complex processes by which stories circulate, collective memories form, “other”-oriented narratives are shaped and reproduced, and the lived social realities of the refugees and the Turkana diverge. The analysis suggests that the refugee and Turkana groups largely coexist in a dynamic landscape where distrust that has been reinforced by numerous narratives and experi- ences of wrongdoing and aggression can occasionally explode into violence. However, the analyses also suggest that violence and distrust is mitigated by individual and group experiences of cooperation, collaboration, and mutual benefit, largely through the exchange of labor, goods, and services, and aided by the presence of the local com- mercial market. Narratives There are numerous narratives about refugees among the Turkana that emerged from interviews conducted for this study. These narratives merge with others about the gov- ernment; the political system; and external factors, such as climate/environment and drought/famine. In an attempt to unpack these factors, narratives have been grouped into subnarratives: (1) refugees are bad, (2) refugees are good, and (3) refugees are ben- eficial; as well as metanarratives: (1) refugees are violent “others” or foreign usurpers, and (2) the Turkana are neglected by the government and discriminated against by “down-Kenyans” and that they are resilient—summed up by the expression “naenda Kenya.” 30 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Subnarrative 1: Refugees are bad. One of the most dominant subnarratives emerging from respondents across the four locations visited in this study is that “refugees are bad.” Most informants either claim to believe this narrative or have heard it repeated in their presence or through interaction networks. There are variations in the Turkana’s negative perceptions of the refugees, but some common themes emerge, including: n Refugees display violence toward the Turkana. According to the Turkana infor- mants, this violence manifests itself both overtly and covertly. Overt forms of vio- lence include: – Open conflict and clashes with weapons that involve large groups of men from refugee and the host communities, remembered through mythicized histories and memories, such as the 2003 small-scale “war”; Shopping for shoes in Hong Kong, Kakuma 4. Interactions at Kakuma 31 – Periodic communal clashes between refugees and hosts when reports or expe- riences of mistreatment by refugees among Turkana men, women, or children lead to small clashes that are usually sorted out through dialogue and external intervention; and – Daily interactions during which Turkana men, women, and children (usually older men and women and young children) experience abuse and are made to leave the camp due to quarrels with or complaints by refugees. Covert means of violence include: – Occasional interactions during which refugees mistreat women and children working for them, with threats to halt further employment; and – Occasional interactions during which usually older men and women might be harmed. n Refugees exploit the Turkana. Exploitation can occur in various ways, including: – Underpayment. n Many Turkana complain that the wages they receive in food or cash are insuf- ficient and unsatisfactory, even while acknowledging it provides enough to survive on a daily basis. n Some respondents complain that refugees try to negotiate prices down to a point that is “insulting” and that the Turkana sellers often have no choice but to sell at the offered prices. – Delayed or nonpayment for goods or services rendered. n Many respondents claim that some refugee groups delay payment for charcoal and firewood delivery as well as for domestic work. n When payments accumulate over time, some refugees refuse to pay the Tur- kana, often chasing them away. 32 Refugee Impacts on Turkana Hosts: A Social Impact Analysis – Overworking. n Respondents mention various occasions when refugees overworked the Tur- kana under their employ, making them carry heavy loads, such as water, food, wood, and fencing material. n Refugees take resources from the Turkana. Accusations of theft and resource extraction take various forms. – Refugees take resources from nongovernmental organizations (NGOs), the United Nations, and the Kenyan government that should either go to or be shared with the Turkana; or – Refugees take land from the Turkana and cause the environmental depletion of Kakuma and the decline of herding and grazing land. n Refugees disdain the Turkana. Many respondents express their sense that the refu- gees think the Turkana are inferior and treat them with varying degrees of disdain. – Teasing. Refugees make fun of Turkana lifestyles, behaviors, and conditions. – Indifference. Many refugees treat the Turkana as though they are invisible and voiceless, ignoring their requests for support or money and often refusing to even acknowledge them. n Refugees bring crime, drugs, and alcohol into Turkana. – Crime. Refugees are involved in stealing. – Drugs and alcohol. Refugees trade in drugs and brew alcohol, which can be abused by the Turkana. Respondents from the five sites offer these narratives in decreasing order of frequency: most respondents refer to the refugees as violent exploiters and usurpers; only a few respondents from Kakuma mention more specific issues, such as disdain, discrimina- tion, crime, and substance abuse. General narratives about refugees as violent “others” or foreign usurpers are common across the sites. The Turkana of Kakuma offer more specific and detailed narratives, possibly because they interact more directly and fre- quently with refugees than do the Turkana of Lorengo, Lokichoggio, and Lorugum, 4. Interactions at Kakuma 33 who more likely hear about these perspectives indirectly and therefore generate rather generic narratives, unburdened by detail or experience. The narratives of the Turkana of Kakuma are nuanced, reflecting the complex nature of their interactions, but their nar- ratives also vary based on the ethnicity of the refugee and their expectations regarding the outcomes of interactions. According to many study informants: n Ethiopians and Darfuris are good, peaceful, will pay cash, and are fair; n Somalis are peaceful and hire more Turkana than any other group, but they fre- quently do not pay on time and break promises; n Burundians, Congolese, and Rwandans are good and peaceful, and they usually pay the agreed-on amount on time; and n The South Sudanese are like the Turkana—they like to drink and are violent. Subnarrative 2: Refugees are good (people). This narrative is not commonly reported by the respondents across the five sites, but is reported extensively and in detail by the respondents at Kakuma and Lorengo. Primar- ily, the ethnographic interview data suggest that significant numbers of the Turkana of Kakuma—and to some extent of Lorengo—report experiences of friendship and part- nerships with refugees. These reports extend beyond the narrative of the refugees ben- efitting the Turkana (see next section—subnarrative 3) through employment, exchange, or the presence of relief and development organizations. This refugee-as-good narrative can be divided into three categories: n Refugee as friend and neighbor. Many of the respondents in Kakuma Many report that refugees are report that the refugees provide help in the form of friendship and generous people who help the support networks that can be drawn on in times of need, including Turkana with gifts of food or cash cash and food transfers and other types of assistance, such as over- even if there is no exchange of night stays and firewood or charcoal storage, as examples. labor, services, or products. n Refugee as helper. Many respondents report that refugees are gen- erous people who help the Turkana with gifts of food or cash even if there is no exchange of labor, services, or products. n Refugee as donor. Many respondents claim that refugees often give money to the Turkana who beg or who seem to be in dire need. Even though the amounts are small, the refugees are acknowledged for their generosity. 34 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Similar to the analysis of “refugees are bad,” in Lokichoggio, and Lorugum, the nar- rative of “refugees are good” is uncommon, but the Turkana of Kakuma and Lorengo discuss it in detail—it is the dominant alternative to the “refugees-are-bad” narrative. It can again be argued that a greater number of interactions can generate more detailed and nuanced positive narratives of refugees among the Turkana living in close proxim- ity to the refugees in Kakuma. Subnarrative 3: Refugees are beneficial. The narrative that refugees are beneficial—that their presence brings benefits to the Turkana—is common across the five sites, especially in Kakuma and Lorengo, and to a lesser extent in Lokichoggio. Respondents acknowledged that the presence and activities of the refugees, such as in the marketplace and by providing employment, fill in gaps in development and relief resources provided by the Kenyan government and the other NGOs and relief organizations to the locals. The respondents suggest that they also benefit from the refugee presence through services offered to refugees that they themselves are also entitled to, including health and medical facilities at the main hospital. Respondents cite a number of specific ways that the Turkana benefit from the presence of refugees, including the following: n Refugees are good for trade. – Consumers of Turkana products. Refugees are the primary consumers of prod- ucts made and sold by the Turkana, including fencing material, construction poles, firewood, charcoal, cereals, and livestock. – Turkana as consumers of refugee products. The Turkana are able to access mar- kets and goods at lower prices inside the refugee camp than they can at other sites, including Kakuma town. – Refugees as sources of credit. The Turkana are sometimes offered credit by refu- gee-owned shops, enabling them to purchase goods that they could not otherwise access. n Refugees as employers. – Domestic work. Refugees provide the Turkana—mainly women and children— with work in return for food and/or cash, benefiting even the least educated with employment opportunities. 4. Interactions at Kakuma 35 – Portage. In return for food and/or cash, refugees hire the Turkana to carry water from the boreholes and relief food from the distribution points to refugee homes. – Home construction. Turkana men and women are hired by refugees to construct fences, build houses, dig ditches, and engage in other construction activities. n Distribution of relief food. Apart from employing the Turkana to carry relief food, respondents also mention that the presence of a black market where refugees sell their relief food results in cheaper prices at the refugee camp and in town markets for food grains and cereals, such as maize and sorghum. n The presence of UNHCR, the International Rescue Committee, and other NGOs. The respondents acknowledge that the refugee camp is the primary reason for the presence of international development and relief organizations, including the UNHCR, as well as the Kenyan government. Reported benefits include: – The additional personnel in the District Commissioner’s Office due to the larger population of Western Turkana District increases accountability and improves engagement with the people. – Relief groups and programs, such as the Hunger Safety Net Programme, provides food and other needs to the Turkana during times of crisis. – The International Rescue Committee’s main hospital provides free health care to the Turkana, including diagnosis and treatment. Despite reported shortcomings in the provisioning of health care, most respondents note that the presence of the hospital is a primary and necessary benefit accrued to them mainly because of the presence of refugees. The narrative of refugees as beneficial to the Turkana can also be parsed along the factor of distance between a respondent’s residence and the camp and the number of interac- tions that the respondent has had with refugees. The Turkana of Kakuma and Lorengo benefit from the presence of refugees more than the Turkana of Lokichoggio, Lorugum, and Lodwar. Interestingly, however, the Turkana of Lokichoggio refer to the refugees as “beneficial” much more often than do those of Lorugum and Lodwar. It is possible that the observable decline in the well-being of the Turkana at Lokichoggio began when the UN and associated relief and development organizations left for Juba without leaving behind an alternative source of jobs, employment, and support for the locals. Therefore, the Turkana of Lokichoggio hold largely negative perceptions of refugees, but at the 36 Refugee Impacts on Turkana Hosts: A Social Impact Analysis same time, as discussed in more detail below, acknowledge that the presence of the refugees is linked to benefits received from associated relief and development agen- cies providing services to the refugees: as a result of agreements between the Kenyan government and UNHCR, some of these services must be provided to the local host community as well. Fathers, brothers, and sons 4. Interactions at Kakuma 37 Metanarrative 1: Refugees are violent “others” and foreign usurpers. An analysis of the interface between subnarratives 1–3 suggest that, across the county, the Turkana share a broad metanarrative of refugees as the violent “others” and foreign usurpers of Turkana land and resources. However, as pointed out earlier, this metanar- rative is quite general and largely without nuance, mainly comprising statements about dangers posed by the presence of refugees. As expected, this larger metanarrative is transmitted in its more general form across the county, and the further the distance between Turkana residents and the refugee camp, the fewer the interactions between groups and the less nuanced the reproduced version of the narrative. The Turkana residents of Kakuma and Lorengo have a nuanced and more complex metanarrative about refugees. The perceptions of refugees among most respondents are still subsumed in the narrative of violent “other” or foreign usurper, but they are also emphatic that they benefit from the presence of the refugees and that refugees are people who range from “good” to “bad” and should be judged on an individual basis. Among the Turkana of Kakuma and Lorengo, the metanarrative of the refugee as violent “other” or foreign usurper is mitigated by interactions between individual members of the host and refugee communities through pathways of social relation- ships, such as friendss, neighbors, and marital alliances, as well as through the exchange of labor, services, products, and even ideas. Metanarrative 2: “Naenda Kenya”—We are marginalized, excluded, and discriminated against. The second metanarrative among the Turkana relates to a sense of marginalization, exclusion, and discrimination from the overall political and social systems and struc- ture of Kenyan society and governance. Aspects of the narrative include: n The Turkana were systematically marginalized by the British and subsequently by the Kenyan government through: – Punitive military action and confiscation of livestock by the British; – The closed district administration and resulting isolation and policed supervision of the Turkana as individuals and as a society (1930–86); – A lack of purposeful development in Turkana County; and – Ongoing narratives among policy makers and political elites about the inability and/or unwillingness of the Turkana to participate in ongoing development efforts. 38 Refugee Impacts on Turkana Hosts: A Social Impact Analysis n The Turkana are excluded from mainstream Kenyan society. “Down-Kenyans” dis- criminate against the Turkana by: – Excluding them from jobs given to down-Kenyans even if they are qualified, unless they are low-paying, menial, or security-related; and – Socially discriminating against the Turkana, even as individuals, treating them poorly through exclusion and marginalization in Kenya and elsewhere. n The Turkana see their primary problem as neglect by the government. Respondents report that: – The government does not care about the Turkana and that even county govern- ment politicians are not concerned about their well-being. – The political experience is an exercise in futility, where politicians come, ask for votes, make promises, and then disappear—only to reappear during the next elec- tion cycle. n The systemic corruption across the political landscape makes it impossible for the benefits of any intervention to effectively and meaningfully reach the Turkana people themselves. This includes planned activities and changes such as: – Development efforts by the UN and other international bodies. Most respon- dents report that development efforts will be successful only if external actors directly speak with the people, going beyond the local leaders and politicians. – Natural resource extraction. Most respondents express cautious optimism that recent discoveries in oil and gas in the Lokichar area and the water aquifer in the Lotikipi plains will benefit the people of Turkana. Most think that any benefits from these industries to the local people will be minimal, such as: n Jobs for the educated. This narrative is given primarily by respondents who complain that they will be excluded from benefits because they are not edu- cated and that they will be relegated to menial labor in similar ways that they already are. n Menial jobs for the Turkana. This narrative is primarily proffered by the edu- cated Turkana respondents who feel that even the educated among the Tur- kana would be hired as sweepers, cleaners, and security guards, and for other 4. Interactions at Kakuma 39 menial low-paying jobs. They would not be given higher-level jobs despite their qualifications. – Devolution. Many respondents offer a mixed narrative about devolution, with most expressing hope tempered by a sense of cynicism about the government and its ability to deliver on the stated promises of transparency and accountability. Responses regarding devolution are of two main types: n If done correctly, devolution will help because the local government would have more autonomy and access to resources. The systems of accountability and transparency would hopefully mitigate the graft and corruption, which is seen as endemic in the political system. Respondents report major concerns regarding the ability of n Even if the central government gives money to the county, the government and the political without a major change in the political system, the money system to transfer the benefits of might not be spent on the people. any intervention to the people. Respondents report major concerns regarding the ability of the gov- ernment and the political system to transfer the benefits of any intervention to the people—whether it be natural resource extraction, development projects, or devolu- tion—but this should be contextualized in the broader narrative about corruption and graft. The main points that emerge from the analysis of the various subnarratives and meta- narratives include: n Ideas and attitudes about the refugees are more nuanced among the Turkana of Kakuma and Lorengo than with the other sites, including the coexistence of positive and negative narratives that have been and continue to be shaped by proximity to the refugee camp and frequency of interactions between the groups; n The Turkana of Lokichoggio hold nuanced ideas about refugees, ranging from the idea that refugees bring benefits to refugees and are violent usurpers, likely resulting from their experience with the sudden depletion of development and relief services when the relevant agencies and NGOs departed after 2008; n Among the Turkana of Lodwar and Lorugum, perceptions of the refugees are more negative: they are more likely to subscribe to the general metanarrative of refuge as violent “other” and foreign usurper without being tempered by coterminous ideas of refugees bringing socioeconomic or other benefits; and 40 Refugee Impacts on Turkana Hosts: A Social Impact Analysis n The Turkana share perspectives of neglect and marginalization by their government and the rest of Kenya under the metanarrative of “naenda Kenya,” which is further tempered by the history of political violence toward and social exclusion of the Tur- kana within and outside Turkana County. This metanarrative is also shaped by issues of corruption and graft that contribute to the cynicism and negative perceptions regarding any intervention where local, regional, or national politicians could play a deciding or determining role. These narratives usually shape and are shaped by interactions and external factors, as explored in the following two sections. Interactions and Narratives Interactions between refugees and the Turkana of Kakuma are shaped largely by the narratives and experiences of satisfaction and dignity or dissatisfaction and indignity emerging from them (figure 4.1a–f). Interactions between refugee and host communi- ties can be divided into two main types: n Social interactions. These fall into two subcategories: – Benign, solidarity-building interactions. These include friendship, marital alli- ances, and networks of support between interacting individuals and families that rely on the Turkana concept of bond-friendship and the refugees’ own cultural logic of friendships and alliances. These interactions often take place in churches or other religious forums; through community organization groups; and at music festivals, weddings, cafes, teashops, and restaurants that are accessed by individu- als from both the host and the refugee communities. – Violent or fractious interactions. In these types of interactions, the Turkana feel threatened, exploited, or are physically and verbally attacked over disagreements regarding social or economic exchange, a cultural misunderstanding, or intoxi- cation/substance abuse. These interactions often take place at social gatherings in public venues, where a public argument can rapidly escalate into a commu- nal conflict. Even more often, however, these types of interactions occur within households due to differing cultural logic or assumptions about payment, refusal, work conditions, and entitlement to returns. 4. Interactions at Kakuma 41 Figure 4.1. Perceptions of Interactions with Refugees Among the Turkana by Age and Gender a. Turkana Women, ages 18–35 (n=28) b. Turkana Men, ages 18–35 (n=18) Early marriage/pregnancy Negative Child abuse Negative Family/marital issues Positive Spread disease/bring disease Positive NGOs don’t care about Turkana NGO/UN (negative) Refugees cause unemployment NGO/UN (negative) Refugees get resources Turkana should get NGO/UN (positive) NGOs don’t care about Turkana NGO/UN (positive) Conflict (general) Conflict (general) Refugees kill Turkana Refugees kill Turkana Refugees poison Turkana Refugees poison Turkana Refugees sexually exploit Turkana Refugees rob/steal from Turkana Refugees rob/steal from Turkana Refugees cheat Turkana in trade Refugees cheat Turkana in trade Refugees harass Turkana Refugees harass Turkana Insecurity (general) Insecurity (general) Increase in crime Prostitution Direct economic assistance from refugees Increase in youth/child crime/drugs/alcohol Refugees as friends/neighbors Increase/introduction of drugs/alcohol Refugees bring resources/services Environmental depletion Refugees good for trade Direct economic assistance from refugees No impact Refugees as friends/neighbors Refugees bring resources/services 0 5 10 15 Refugees good for trade 0 1 2 3 4 5 6 7 c. Turkana Women, ages 36–55 (n=30) d. Turkana Men, ages 36–55 (n=36) Spoiled children Negative Divorce Negative Child abuse Positive Child abuse Early marriage/pregnancy Positive Family/marital issues NGO/UN (negative) Early marriage/pregnancy NGO/UN (negative) Spread disease/bring disease NGO/UN (positive) Spread disease/bring disease NGO/UN (positive) Refugees cause school dropouts NGOs don’t care about Turkana Refugees cause unemployment Refugees get resources Turkana should get NGOs don’t care about Turkana Refugees get too many resources/services Refugees get resources Turkana should get Conflict (general) Refugees get too many resources/services Refugees kill Turkana Refugees kill Turkana Refugees poison Turkana Refugees poison Turkana Refugees sexually exploit Turkana Refugees sexually exploit Turkana Refugees rob/steal from Turkana Refugees rob/steal from Turkana Refugees abduct Turkana children Refugees cheat Turkana in trade Refugees cheat Turkana in trade Refugees harass Turkana Refugees harass Turkana Insecurity (general) Insecurity (general) Prostitution Prostitution Increase in youth/child crime/drugs/alcohol Livestock theft Increase/introduction of drugs/alcohol Increase in youth/child crime/drugs/alcohol Increase in crime Increase/introduction of drugs/alcohol Refugees as friends/neighbors Increase in crime Refugees bring resources/services Taking land from Turkana Refugees good for trade Environmental depletion 0 2 4 6 8 Direct economic assistance from refugees Refugees as friends/neighbors Refugees bring resources/services Refugees good for trade No information 0 2 4 6 8 10 12 e. Turkana Women, ages 56–81 (n=23) f. Turkana Men, ages 56–81 (n=15) Refugees abduct Turkana children/ Negative Spoiled children human trafficking Positive Child abuse Negative Spread disease/bring disease Positive NGO/UN (negative) Refugees cheat Turkana in trade Refugees cause school dropouts NGO/UN (positive) NGOs don’t care about Turkana NGO/UN (negative) Refugees get resources Turkana should get NGO/UN (positive) Prostitution Conflict (general) Refugees kill Turkana Refugees as friends/neighbors Refugees sexually exploit Turkana Refugees rob/steal from Turkana Refugees good for trade Refugees cheat Turkana in trade Refugees harass Turkana Prostitution No impact Livestock theft Increase/introduction of drugs/alcohol No information Environmental depletion Refugees as friends/neighbors Refugees good for trade 0 2 4 6 8 10 No impact No information 0 1 2 3 4 5 Source: World Bank survey 2015. 42 Refugee Impacts on Turkana Hosts: A Social Impact Analysis n Economic interactions. They include: – Exchange of labor. In this circumstance, the Turkana work directly for the refu- gees as domestic servants, in construction, or in portage. These transactions often get conflated with social interactions and hence can be benign if both parties in the exchange can maintain dignity and satisfaction; but they can become violent over a misunderstanding or with accusations of exploitation. – Exchange of products. In these interactions, the Turkana provide needed items to the refugees, such as fuelwood, meat, and other livestock products. In turn, they are able to access markets established by the refugees. These interactions allow for the flow of goods, cash, and food between refugees and hosts, but dis- agreements over payments, accusations of theft, and the humiliation associated with bargaining can lead to quarrels and conflict. Turkana children with Refugee neighbors 4. Interactions at Kakuma 43 As described above, many interactions are shaped by social narratives and ongoing interactions that reinforce or modify how various narratives are manifested and used by Turkana individuals and groups. Hence, the metanarrative of the refugee as vio- lent “other” acts as a larger cautionary tale within which the Turkana of Kakuma and Lorengo can negotiate what they might find to be a dangerous space with the potential to be treated with indignity. Their caution is tempered by their dependence on the camp for social and economic interactions and their understanding that the refugees bring direct and indirect benefits to the Turkana and their families. It should be noted that the refugees are only one of a number of concerns among the Turkana. Other narratives emerge from the larger subsistence and political economies of the Turkana in which they present themselves as the lords of their domain, as resil- ient and successful people in the face of external and uncontrollable factors such as drought, famine, climate change, and other environmental issues. These narratives are described in the next section. External Factors and Narratives of Suffering, Resilience, and Dignity The Turkana, who are predominantly subsistence pastoralists, are extremely concerned with the availability of grazing land and water, factors that depend on annual and sea- sonal rainfall and precipitation regimes. As explained earlier, the Turkana occupied the western Lake Turkana basin starting from southeastern Uganda and the Karamoja escarpment in the 16th century CE and then expanding out in the 19th century CE to areas that include present-day Turkana County and beyond into other parts of Kenya, Uganda, South Sudan, and Ethiopia. During this latter period, the Turkana became pastoralists because it enhanced one’s status from only pursuing subsistence activi- ties. Historical and archaeological records suggest that the Turkana followed diversi- fied subsistence pathways, including hunting, foraging, and agriculture in addition to pastoralism. These narratives of a grand past are paralleled by other historical narratives of margin- alization and exclusion by the British and Kenyan governments as well as narratives of resilience in the face of environmental and climatic changes, especially droughts and famines. The Turkana pride themselves on having successfully managed pastoralism and other subsistence and status-building activities for centuries, and they have devel- oped an extensive suite of local and regional narratives that combine stories of suffering and resilience in the face of drought and famine. Memorized or recounted stories are told of the suffering of the respondents themselves or their family members during these famines and droughts, followed by stories of coping and resilience that include 44 Refugee Impacts on Turkana Hosts: A Social Impact Analysis strategies of migration, foraging, and cooking famine foods as well as grief and mourn- ing. Respondents also relate narratives of birth, marriage, and regeneration of livestock and families during and after the famines. The narratives of famine and drought feed into the larger narratives of environmental and climatic change, where the respondents themselves report observations of both increasing frequencies of drought and lethality of famines, especially to livestock. These narratives occur more frequently among the Turkana of Lorengo and Lokichoggio than among those of Lorugum and Kakuma. It could be argued that the generation and reproduction of the famine narratives are directly related to the decline of regional and local networks among the Turkana. Perceptions as Outcomes The various interactions noted in the previous section are combined into categories of perceptions: general negative (nonviolent), negative (violent “other”), and positive (good and/or beneficial). General Negative Perceptions Levels of negativity in the Turkana’s perceptions of refugees, excluding reports of vio- lence and mistreatment, are generally equal across the four locations studied in the host community (figure 4.2). There is a statistically significant difference between Kakuma and Lorugum in this metric at the 95 percent confidence level (X2=5.08; p=0.024), but not between Kakuma and Lorengo, with the second greatest difference in proportion (X2=2.79; p=0.095). There is also no statistically sig- nificant difference between Lorengo and Lorugum Figure 4.2. Proportion of Host Community (X2=0.201; p=0.654). These results suggest that a Reporting Negative (Nonviolent) Refugee Impacts similar proportion of people across Turkana County 35 33 have negative opinions about the impact of refugees. 31 30 Notably, however, of the 38 respondents in Lorugum 25 23 who had opinions about refugees, six report vio- 21 lent behavior or mistreatment, and 31 report other Percent 20 15 negative impacts of the refugees, indicating that the 10 limited information that makes its way to Lorugum 5 about refugees is overwhelmingly negative. 0 Kakuma Lorengo Lokichoggio Lorugum Importantly, negative opinions are not exclu- Source: World Bank survey 2015. sively directed at the refugees themselves. Most 4. Interactions at Kakuma 45 interviewed respondents suggest that the Turkana feel the most animosity toward the political system and even toward UNHCR and other NGOs providing aid to refugees in Kakuma, such as the Lutheran World Federation and the World Food Programme (WFP). The Turkana often feel that these agencies are unjust in their distribution of resources because the host community is also in need of aid and economic develop- ment. They frequently cite the long history of marginalization they have experienced as well as their traditional ownership of the land on which the refugee camp is located to demonstrate that they deserve attention from international actors in the region. They perceive these organizations to have vast resources at their disposal, and they tend to believe that the decision to prioritize refugees over the Turkana is rooted in malice and discrimination. Perceptions of Refugees as Violent “Others” In response to the question about the impact of refugees on their lives, 246 of the 601 Turkana interviewed (41 percent) report that they experience some combination of mistreatment, economic malfeasance, and violence at the hands of refugees when they enter the camp to trade (figure 4.3). Notably, the proportion of Turkana respondents reporting these behaviors varies significantly by location. In Kakuma, which is adjacent to the refugee camp, 50 of 150 respondents (33 percent) report refugee violence and mistreatment. Interestingly, an even higher proportion of respondents in both Lokichoggio (58 percent) and Lorengo (68 percent) report these behaviors. While there is no statistically significant difference between Lokichoggio and Lorengo with regard to the proportion of reports of refugee violence at the 95 percent confidence level (Χ2=3.23, p=0.072), both of these proportions are Figure 4.3. Proportion of Host Community significantly higher than in Kakuma (comparing Members Reporting Refugee Violence or Lokichoggio, the lower of the two proportions with Mistreatment Kakuma: X2=17.41; p<0.001). Although the limited 80 69 information about refugees trickling to Lorugum 70 60 58 is negative, all six respondents at this location who 50 spoke of refugees report mistreatment or violence Percent 40 33 toward the Turkana. 30 20 Despite the ubiquity of stories about refugee vio- 10 4 lence and mistreatment, none of the Turkana inter- 0 Kakuma Lorengo Lokichoggio Lorugum viewed in all four sites claim first-hand experience of this kind, and only two report knowing someone who Source: World Bank survey 2015. 46 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Figure 4.4. Proportion of Host Community had been victimized by refugees. One of these reports Members Reporting Positive Refugee Impacts came from a woman in Lorengo who said that her neighbor’s family all fell ill after eating food pur- 80 70 49 chased in the refugee camp. She attributes the illness 60 45 to deliberate poisoning, though no evidence was 50 produced. When asked, most participants admit that Percent 40 the stories heard about refugee violence are rumors. 30 16 20 Positive Perceptions 10 0 0 Kakuma Lorengo Lokichoggio Lorugum A slightly different pattern emerges with respect to perceptions that refugees have some positive Source: World Bank survey 2015. impact on the host community. As distance from the refugee camp grows, the proportion of people who believe that refugees have any positive impact decreases (figure 4.4). In Kakuma, there was actually a higher proportion of respon- dents who note positive refugee impacts (49 percent) than violence or mistreatment. In rural Lorengo, 45 percent of respondents report positive refugee impacts; statistically, these proportions are not significantly different from one another (X2=0.482; p=0.488). In both Kakuma and rural Lorengo, a high proportion of these positive reports involve trading and other economic opportunities offered by refugees. Specifically, several respondents in both Kakuma and Lorengo report having friendships, or at least friendly relationships, with refugees. Others report that refugees offer employment opportuni- ties—primarily domestic work in refugee homes—and a few note that the host com- munity sometimes has access to aid and services from NGOs serving refugees. Similar to the overall positive sentiments, these specific reports decline with increasing distance from the camp. In Lorugum, far from Kakuma, no one reports benefits to the host com- munity because of the refugees or any other positive sentiments about them. 5 Measuring Impacts Measuring Nutritional and Psychosocial Well-Being Among the Turkana As described in the methodology section, part of the social impact analysis was a large- scale survey of the Turkana residents of Kakuma, Lorugum, Lorengo, and Lokichoggio, which collected data on demographic factors (age, gender, number of livestock, and number of children); anthropometric factors (height, weight, and skinfold thickness); and psychosocial stress (listing of various causes of stress). These data were collected from 600 individuals—300 male and 300 female, with 150 respondents from each of the four sites. The height, weight, and skinfold measurements enabled the calculation of two anthropometric variables that are considered significant indicators of nutritional well-being among human populations: body mass index (BMI) and sum of skinfolds (SSF). Both of these enable the calculation of the body fat percentage (BFP) of a human or other living being, which is the total mass of fat divided by total body mass. Body fat includes essential body fat and stored body fat. Essential body fat is necessary to main- tain life and reproductive functions. While BMI is useful as an approximate indicator of well-being, SSF is seen as a more accurate indicator of BFP, especially conducted among ethnically and nutritionally similar groups. It is based on a skinfolds test, also known as a pinch test, whereby a pinch of skin is precisely measured by calipers at several standardized points on the body to determine the subcutaneous fat layer thickness. SSF is a reliable measure of an individual’s energetic status because it enables reliable and accurate measures of BFP (Campbell et al. 2005). The data were analyzed using various procedures to: n Establish a baseline for the nutritional and psychosocial status of the Turkana at Kakuma, Lorugum, Lokichoggio, and Lorengo; n Examine possible impacts of the refugee presence at Kakuma on the host community by analyzing the BMI, SSF, BFP, and psychosocial stress profiles within and between the populations of the four sites; and 47 48 Refugee Impacts on Turkana Hosts: A Social Impact Analysis n Understand how gender, age, and location might affect the physical and psychosocial well-being of the residents of the four sites in Turkana County. Additional caveats to this study include the following: n Sedentary Turkana men would show higher measures of BMI, SSF, and BFP than nomadic or pastoral Turkana men. However, all the women in this study followed a more sedentary/settled lifestyle. The women in pastoral households were more sedentary, raising children and young animals, making charcoal, and collecting firewood. Some of the women in Kakuma and Lorugum, and to a lesser extent Lokichoggio, also mentioned office work, teaching, and domestic service. n SSF may vary based on body type and diet. Although this could be due to greater access to cereals and carbohydrates at Kakuma as opposed to the more meat- and milk-based diet among pastoralists of the more rural areas of Lorengo, the Turkana pastoralists are not as cereal-averse as usually assumed. In both their daily lives and their mythic tradition, plant-based foods, including sorghum, are regarded as staples, but cultural adaptations have taken place where older people eschew cereals so that younger people can get access, especially during droughts and famines—a practice confirmed by study respondents across all sites (Lamphear 1988; Wright et al. 2015). Furthermore, the Turkana of today include maize and beans in a large por- tion of their diet, a trend especially noted after the vast inflow of cereal-based relief food that arrived following the famines of 1990–92 (called Lokwakoyo/Akalkal), 1999–2000 (called Logara/Empopo), and 2011–12 (Hunger Safety Net Programme), as well as during droughts and the periods between the famines. As Bush (1995: 249) noted, when traditional foods were depleted, the Turkana grew to “increasingly rely on commercial food (maize and beans) and, to a lesser extent, on wild fruits” leading to “higher consumption of [not local] food” and a greater “reliance on the market and, in turn, on agricultural production in distant farming regions.” We observed that sorghum, maize, and beans were staple foods across the four sites of this study. Therefore, differences in physical indicators (BMI, SSF, and BFP) between Kakuma and the other sites could be due to enhanced access to larger quantities of available cereals at Kakuma. n Regarding water, sanitation, and hygiene, it is clear from various studies that expo- sure to environments with high levels of fecal contamination is correlated with chronic malnutrition and undernourishment caused by recurring diarrhea and pos- sibly tropical or environmental enteropathy (Humphery 2009; Mahmud and Mboya 2016). We would expect all the sites to have high rates of malnutrition related to water, sanitation, and hygiene due to the sharing of water sources with livestock 5. Measuring Impacts 49 and other domesticated animals, such as riverbed holes, rivers, lakes, and troughs from boreholes. Significant impacts to the population due to a lack of water, sanita- tion, and hygiene are likely in Kakuma because, in addition to having the highest population density among the sites (>2,000 persons/square kilometer), most of its host population is still meeting their personal and livestock water needs using holes dug in the bed of the Tarach River and local water holes, and the use of open spaces as latrines persists due to a paucity of pit latrines. As the results suggest, while water, sanitation, and hygiene may have an impact, the nutritional and energy status of the Turkana in Kakuma is higher than for those in Lorengo or Lokichoggio, most likely due to their greater access to energy-rich foods such as cereals. Body Mass Index Nutritional data on height and weight (mass) measurements were used to calculate BMI [body mass (kg)/(height/m)2]. The results sug- Table 5.1. BMI Across Four Locations gest that the average BMI for both men and women was higher in Location Men Women Kakuma and Lorugum than at Lorengo or Lokichoggio (table 5.1). Kakuma 18.24 19.27 Lokichoggio 18.05 19.14 A series of tests were performed to examine if there were any dif- Lorengo 18.21 18.63 ferences in BMI after disaggregating the populations of the four Lorugum 18.29 19.61 locations by age and gender (see table A.1). The differences in Source: World Bank survey 2015. BMI between the men at the four locations were not statistically significant. However, as table 5.2a demonstrates, the analysis of the BMI from all the respondents—300 men and 300 women—suggest that both are underweight. In addition, the analysis of the proportion of population who are thin/undernourished, normal, or overweight/obese was compared with the results of the Kenya Demographic Health Survey (KDHS 2014). The results of both surveys are comparable, with a simi- lar distribution of BMI for Turkana women across the three categories (table 5.2b). A similar breakdown was not found for men, neither for the national aggregates nor for Turkana-specific results, signifying a severely gendered gap in data. Nevertheless, the results suggest that 45.3 percent of women and 58.9 percent of men in Turkana are highly vulnerable to malnutrition caused by food insecurity. The BMI data were further disaggregated by age—young adult (18–35), middle-aged (36–55), and older adults (>56)—and then by age and gender groups. The BMI varia- tion by age and gender are shown in tables 5.3a and 5.3b. 50 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Table 5.2a. Average BMI of Population in Turkana Compared with the World Health Organization’s Threshold for Malnourishment/Undernourishment Average Standard WHO BMI Deviation Threshold Women 19.2 2.7 BMI < 19.5 Men 18.2 2.0 BMI < 18.5 Source: World Bank survey 2015. Table 5.2b. Average BMI by Age and Gender Group Percent Percent Thin/ Percent Overweight/ Undernourished Normal Obese BMI (<18.5) (18.5–24.9) (≥25) National Women 23.7 8.9 58.3 32.8 Turkana Women (KDHS) 19.4 45.3 50.0 4.7 Turkana Women 19.2 45.4 51.0 3.6 (World Bank survey) National Men 21.6 N.A. N.A. N.A. Turkana Men* 17.7 N.A. N.A. N.A. Men (World Bank survey) 18.2 58.9 40.4 0.7 Note: KDHS = Kenya Demographic and Health Survey. Sources: *Campbell et al. 2005; KDHS 2014; and World Bank survey 2015. As seen in table 5.3a and table A.1 (see appendix A) there is no significant difference in the BMI of the Turkana across all four locations among young adults, the middle-aged, and older adults, but there was a difference in some subpopulations when the BMI was disaggregated by age and gender (table 5b). Young adult females of Lorugum and, to a lesser extent, Kakuma have marginally higher BMIs than do those of Lorengo and Lokichoggio. The BMI analysis within gender and age categories showed small differ- ences across the four locations between women of Lorugum and Lorengo, and when further disaggregated by age and gender, specifically showed small differences between the young women of Kakuma/Lorugum and Lorengo/Lokichoggio. However, the over- all figures are worrying in terms of physical well-being and nutritional profiles of the Turkana because they suggest that undernourishment is a normal state of affairs in Tur- kana County, even in areas where residents enjoy proximity and relatively easy access to relief and development services and resources, such as in Kakuma and Lorugum. There is a greater presence of public services in Kakuma due to the presence of the refugees, 5. Measuring Impacts 51 Table 5.3a. Average BMI by Age Group Middle- Young aged Older Kakuma 19.5 18.8 18.0 Lokichoggio 18.7 18.4 18.7 Lorengo 18.7 18.4 18.4 Lorugum 19.6 18.9 18.3 Source: World Bank survey 2015. Table 5.3b. Average BMI by Age and Gender Group Female Male Young Middle- Older Young Middle- Older Adult Aged Adult Aged Kakuma 20.09 19.61 17.89 18.84 18.02 18.08 Lokichoggio 19.10 19.22 19.22 18.36 17.65 18.08 Lorengo 18.41 18.47 19.15 18.91 18.39 17.62 Lorugum 20.37 19.18 19.26 18.85 18.67 17.39 Source: World Bank survey 2015. and Lorugum has benefited from almost 20 years of consistent attention and invest- ment due to the Turkwel Hydroelectric Power Station, which has enabled engaging in agriculture and more sustainable pastoralism. A variety of studies suggest that BMI must be used in conjunction with other indicators of nutrition because height and body mass (weight) do not accurately account for body shape, body fat, or populations that tend toward particular body types, such as tall and thin or short and stocky. The Turkana, like many East African pastoral groups, tend to be tall and thin with low percentages of body fat. BMI measures often overestimate measurements for taller individuals by 10 percent. But even when the overestimation of BMI is factored in, Turkana respondents demonstrate high levels of less than optimal or only minimal healthy nutritional profiles. Sum of Skinfolds Responding to concerns about the accuracy of using BMI alone to test nutrition levels in a population, health researchers and professionals commonly use another variable— sum of skinfolds (SSF). Results of the analysis of all skinfold measurements among respondents from the four sites are shown in figures 5.1a, 5.1b, and 5.1c. 52 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Figure 5.1a. Average SSF for Men and Figure 5.1b. Average SSF for Young, Women across Turkana County Middle-aged, and Older Women Across Turkana County 35 Women Men 35 Young adult Skinfold thickness (mm) 30 Middle-aged adult Skinfold thickness (mm) 30 Older adult 25 25 20 20 15 15 10 Lorengo Lokichoggio Lorugum Kakuma 10 Lorengo Lokichoggio Lorugum Kakuma Figure 5.1c. Average SSF for Young, Middle-aged, and Older Men across Turkana County The SSF results were subjected to the same series of tests as BMI (see table A.1 in appendix A) 35 Young adult Middle-aged adult showing significant differences between the aver- Skinfold thickness (mm) 30 Older adult age SSFs across the four sites. The average SSF of 25 both men and women at Kakuma and Lorugum are higher than the average SSF at Lorengo and 20 Lokichoggio, similar to the findings in the BMI 15 analysis. 10 Lorengo Lokichoggio Lorugum Kakuma However, as shown in tables 5.4a and 5.4b, the Source: World Bank survey 2015. results also suggest that while Kakuma and Loru- gum show higher average SSF values than Lorengo and Lokichoggio, there is far greater variation of the SSF measurement among both men and women within the populations of these locations than there is at either Lorengo or Lokichoggio. Kakuma and Lorugum reveal a far greater variation in SSF measure- ments for men and women than the distribution of SSF in Lorengo and Lokichoggio. Using analysis of variance (ANOVA), the differences between and within sites was found to be highly significant (p<0.001), especially seen in combination with the average SSF measurements for men and women across all of the locations. The mean SSF values by gender and location suggest that the Turkana of Lorengo and Lokichoggio dispropor- tionately suffer from undernourishment and lack of access to adequate nutrition, but the lack of access to nutrition is felt equally across societies, indicating the more widespread distribution of poverty at these sites. 5. Measuring Impacts 53 The higher average of the SSF values at Kakuma and Lorugum suggest that Turkana residents in these locations have greater nutritional security and health than those of Lokichoggio and Lorengo. While SSF findings for Lorugum suggest that the presence of consistent development due to the Turkwel Hydroelectric Power Station has led to greater food security, they also suggest increased inequality and even inequity through the generation of economic classes. Similar conclusions could be drawn from the SSF findings at Kakuma, which suggest that the Turkana of Kakuma have greater access to food and hence nutritional security because of the presence of the camp. These findings confirm one of the expectations of this Table 5.4a. Average SSF Measurements and study: that the greater opportunities for labor and Variations in SSF by Location and Gender exchange that emerge from the refugees’ relief and development needs are correlated with enhanced access Average Standard Location Gender SSF Deviation to cash and/or food exchanged for Turkana labor, prod- ucts, and services. However, the analysis also suggests Kakuma Female 30.85 19.82 that these opportunities are contingent on the demand Male 17.02 9.37 of these services and goods and are subject to prices that Lokichoggio Female 23.40 12.68 are controlled by the consumers. The inconsistency of Male 14.29 2.99 work opportunities and cash or food flow is only occa- Lorengo Female 17.57 7.29 sionally mitigated through acts of one-way exchange— Male 14.09 3.05 for example, when the refugees feed the people who Lorugum Female 33.17 20.68 work for them even when no work is available. These Male 16.22 7.45 disparities can be inferred from the greater variation in Source: World Bank survey 2015. SSF seen among both men and women at Kakuma. Table 5.4b. Results of T-Tests Showing Difference in SSF Across Sites Kakuma Lokichoggio Lorengo Lorugum Kakuma — Significant difference— Significant difference— No significant women (p=0.006) women (p=0.001) difference—women (p=0.378) Lokichoggio Significant difference— — Significant difference— Significant difference— men (p=0.009) women (p=0.05) women (p=0.005) Lorengo Significant difference— No significant — Significant difference— men (p=0.005) difference—men women (p=0.001) (p=0.85) Lorugum No significant Significant difference— Significant difference— — difference—men men (p=0.05) men (p=0.001) (p=0.434) Source: World Bank survey 2015. 54 Refugee Impacts on Turkana Hosts: A Social Impact Analysis BMI and SSF data were further analyzed to see if there were differences between popu- lations that interacted with refugees/development-relief activities and the populations who self-identified as pure pastoralists and hence did not interact with the refugees in Kakuma and Lorengo (or with development/NGOs in the case of Lorugum and Lokichoggio). Only 6 out of 150 respondents in Kakuma (3 female, 3 male) fell into the category of nonrefugee/nondevelopment-aided subsistence/livelihood, compared with 28 in Lokichoggio (8 female, 20 male), 140 in Lorengo (65 female, 75 male), and 17 in Lorugum (6 female, 11 male). The results of an analysis of the differences in BMI and SSF for these groups, controlling for age and location, follows: n BMI – Women. The pastoralist women of Kakuma have significantly lower mean BMI (15.65, standard deviation 2.95) than the women of Lokichoggio (18.95, standard deviation 1.71, p=0.023), Lorengo (18.63, standard deviation 2.06, p=0.022), and Lorugum (19.31, standard deviation 2.16, p=0.023). – Men. There are no statistically significant differences in the mean BMI among pastoralist men across all four locations. n SSF – Women. There is no statistically significant difference (p=0.972) between the mean SSF among pastoralist women of Kakuma (15.73, standard deviation 3.84) and Lorengo (17.57, standard deviation 7.29). There are, however, significant dif- ferences between the means of SSF for the women of Kakuma and both Lorugum (45.17, standard deviation 22.39, p=0.004) and Lokichoggio (33.51, standard deviation 26.51, p=0.07). – Men. There are no statistically significant differences in the mean SSF among pas- toralist men across all four locations. These findings suggest that the bulk of the Turkana at Kakuma are engaged with the refugee and/or the relief development sector at Kakuma, with interactions ranging from exchange of labor and goods to begging, all of which provide access to cereal-based foods and hence higher nutritional status. It is particularly worrying that the relatively few men and (especially) women in the Kakuma area who self-identified as “pure” pastoralists have lower BMI and SSF than did self-identified pastoralists in the other 5. Measuring Impacts 55 locations (see appendix A). This may be a result of the low sample size for both men (n=3) and women (n=3) at Kakuma. However, in combination with the ethnographic data on the loss of land, tree cover, water, and livestock market dynamics, the findings stress the need for further study to explore whether the population of Kakuma that is solely dependent on pastoralism is at greater risk and is more vulnerable than the pas- toralists of other parts of Turkana. From a policy approach, the findings suggest a dire need for immediate intervention and aid for the pastoralists of Kakuma in ways that are not dependent on the refugee-host dynamics. On the other hand, the overall findings indicate that engagement with refugee-related activities and exchanges results in higher BMI and particulary SSF for the Turkana of Kakuma. Specifically, results from the BMI and especially the SSF data, controlling for the lower BMI and body fat content expected among pastoralists, suggest that the energy status of the host community of Kakuma is better than that of Lokichoggio or Lorengo and that results are similar to Figure 5.2. Average BFP Across Sites that of Lorogum. a. Women 100 To control for body type (pastoralists typically having lower 90 Skinfold thickness (mm) BFP than agriculturalists or other groups), the SSF measure- 80 70 ments were also converted into body fat percentage (BFP; 60 figure 5.2a and figure 5.2b). These figures show the number of 50 40 men and women in each site whose BFP falls above or below 30 the healthy levels corrected for pastoralists. 20 10 0 Kakuma Lokichoggio Lorengo Lorugum Both figures 5.2a and 5.2b suggest that the women in Kakuma and Lorugum fare better than those in Lorengo and Lokichog- gio, while the men of Kakuma fare marginally better than b. Men men at Lorugum, Lorengo, and Lokichoggio. These analyses 100 90 however, do not show the variation in BFP nor are they con- Skinfold thickness (mm) 80 trolled for age and gender. A regression analysis controlling 70 60 for both age (pastoralists lost BFP as they age) and gender 50 (younger women have higher BFP that declines with subse- 40 30 quent childbirth) shows no differences between the BFP for 20 the men across the four sites. However, there are significant 10 0 differences between the women of Kakuma and Lokichog- Kakuma Lokichoggio Lorengo Lorugum gio (p=0.002), and Kakuma and Lorengo (p=0.0001), but no Above/average Below average difference between Kakuma and Lorugum (p=0.658). (See Source: World Bank survey 2015. appendix B.) 56 Refugee Impacts on Turkana Hosts: A Social Impact Analysis More analysis is needed to ascertain this as fact, including a closer look at diets; water, sanitation, and hygiene; and other factors, but the presence of refugees at Kakuma appears to be filling a general gap, analogous to the development process at Lorugum. It is more highly correlated with a greater average of physical well-being among the host community, possibly resulting from increased access to high energy foods such as cere- als or cash in exchange for goods, services, and labor, as well as other services intended for refugees that are also available to or easily accessed by the host community. Results suggest that the Turkana of Kakuma have better nutritional access and sta- tus than those of Lokichoggio or Lorengo and similar access and status than those at Lorogum, which has experienced consistent development investment for over 20 years. However, the presence of the refugees may also lead to differences in psychosocial stress levels among the host community. Results are described in the next section. Psychosocial Stress The analysis of the psychosocial stressor data first examined the relationship between nutritional status and worries related to eight stressor categories: water, food, health, livestock, environment, economy, refugees, and transportation. To see how these stress- ors are distributed among locations when controlling for sample size within each group and within age and gender categories, the larger set of interactions is combined into four categories: positive, negative, neutral, and systemic. They are reported as the aver- age number of the types of interactions recounted by each respondent and the pro- portion of incidents reported within each of these four categories, disaggregated by age and gender. The results (see figures 5.3a–5.3d) provide a clearer picture and help contextualize the differences and variations among the Turkana of Kakuma in terms of their perceptions of refugees. On average and proportionally, Turkana men report more negative and fewer positive interactions with refugees than women do; they also report more systemic problems than the women (figures 5.3a and 5.3b). As demonstrated in figures 5.3c and 5.3d, young Turkana have more positive interac- tions with refugees, while middle-aged and older Turkana report a relatively lower num- ber of positive experiences with refugees but slightly higher proportions and averages of negative interactions. Middle-aged and older Turkana also report more significant issues with systemic problems and fewer problems with the systemic processes than do the young. Using SSF as a proxy for nutritional status, worry codes (appendix c) were placed into the above stressor categories, which were treated as dummy variables. A series of t-tests were used to assess whether the mean of the sum of skinfolds varied significantly between 5. Measuring Impacts 57 Figure 5.3. Worries Among the Host Community in Kakuma, Disaggregated by Gender and Age b. Proportion of Refugee-related a. Average Number of Refugee-related Worries by Gender Worries by Gender 1.6 100 1.4 90 80 1.2 70 1.0 60 0.8 50 0.6 40 30 0.4 20 0.2 10 0.0 0 Female Male Female Male c. Average Number. of Refugee-related Worries d. Proportion of Refugee-related by Age Worries by Age 1.6 100 1.4 90 80 1.2 70 1.0 60 0.8 50 0.6 40 30 0.4 20 0.2 10 0.0 0 18–35 36–55 56–81 18–35 36–55 56–81 Age Age Positive Interactions Negative Interactions Neutral or No Interactions Systemic Neglect Source: World Bank survey 2015. 58 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Figure 5.4. Density and Variation of those who reported worries in a given category and those who Worries Across Locations did not. Based on these tests, only worries related to health, a. Kakuma livestock, and raiding appear to be correlated with nutritional status (p<0.05). In each of these cases, mean SSF was lower in 1.5 those reporting these worries than those who did not. There is no statistically significant relationship between nutritional 1.0 Density status and stressors related to water, food, the environment, 0.5 the economy, or refugees (p>0.05). There is also no difference in nutritional status between those who reported having no 0 worries and those who reported at least one (p=0.79). 0 2 4 6 8 Total number of worries reported The distribution of the eight stressor categories across the four b. Loki locations is shown in figure 5.4. The mean number of worries 1.5 reported is higher in Kakuma (3.72) than any other location (Lokichoggio=2.99; Lorengo=3.36; Lorugum=2.7). However, 1.0 nutritional status is essentially equal between Kakuma and Density Lorugum for both men and women, and higher in these loca- 0.5 tions than in either Lokichoggio or Lorengo, suggesting that while the Turkana in Kakuma have greater access to food and 0 0 2 4 6 8 other resources than do those in Lokichoggio and Lorengo, the Total number of worries reported refugee camp is associated with greater variation in psychoso- c. Lorengo cial stressors. Kakuma respondents report the greatest range 1.5 of stressors, from 0 to 7 stressors mentioned by individual subjects. An analysis of variance (ANOVA) test shows that the 1.0 reported variance is significant (p<0.001). Density 0.5 Further analysis was conducted using a principal component analysis (PCA) on how different locations are characterized 0 by certain types of worries and stressors. While the worries 0 2 4 6 8 Total number of worries reported reported are discrete data, they were converted to continuous data by representing them as a fraction of the total number of d. Lorugum worries and calculating the frequencies. 1.5 Hunger emerges as a significant variable for almost all of the 1.0 respondents and is the only factor significantly loaded onto Density PC01. Hence, we discount it in the analysis. The actual varia- 0.5 tion of worries across all four locations can be seen by look- ing at PC02 versus PC03 (figure 5.4). Points represent groups 0 0 2 4 6 8 disaggregated by location, age, and gender, and the colored Total number of worries reported Source: World Bank survey 2015. 5. Measuring Impacts 59 Figure 5.5. Principal Components Analysis of Worries Across Sites Notes: L1/red = Kakuma town; L2/black = Lokichoggio; L3/blue = Lorengo; L4/orange = Lorugum. s1 = female; s2 = male. ya = young adult; ma = middle-aged adult; oa = older adult. Source: World Bank survey 2015. circles correspond to gender/location categories. Overall, figure 5.5 shows that location and gender clusters are still quite clear, but age is also a major factor driving the varia- tion in worries. Kakuma. Of all four locations, when age is taken into account, Kakuma has the great- est variation in principal component scores. This variation is primarily loaded onto PC03. Not surprisingly, older adults and young adults vary most dramatically. Among women, thirst (code 11) and unemployment (code 50) appear to be the primary drivers of difference between age-sets; in men, these same worries are driving age variation but are more intensely felt. In both cases, young adults feel these worries more strongly than older adults. For men in particular, the age variation appears to be driven almost exclusively by worry about unemployment: a line drawn between the young adult male and the older adult male points would be almost exactly parallel to the vector associated with unemployment. 60 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Lokichoggio. In Lokichoggio, illness-related worries are driving the variation between older adult women and women of the other age-sets. Thirst and unemployment are the primary drivers of variation between men and women. Among men, variation is loaded almost equally between PC02 and PC03. Young adult men worry most about raids and personal illness; older adult men worry most about livestock illness; and middle-aged adults worry most about thirst and unemployment. Lorengo. Worries about livestock and family illness (codes 43 and 42, respectively); the lack of medical facilities (code 44); and drought (code 10) set Lorengo apart from the other locations, largely due to the nearly complete absence of public health facilities, such as clinics and veterinary services. Age-set variation at Lorengo is loaded almost equally onto PC02 and PC03. Among men, older adults differ the most from young and middle-aged adults, but among women, it is young adults who vary the most from the other two age-sets. Young adult women are most worried about unemployment and poverty (code 51), whereas older women are more concerned with livestock and family illness (codes 43 and 42, respectively). Older adult men vary from younger men primarily based on worries about hunger (code 30) and livestock illness. Lorugum. Lorugum overlaps significantly with Kakuma in terms of worry profiles. In Lorugum, however, the variation of worries is overwhelmingly accounted for by The meat vendors alley 5. Measuring Impacts 61 age rather than gender. Additionally, variation is loaded primarily onto PC03. Unlike Lokichoggio, older adults in Lorugum—men and women—are more worried about raiding, middle-aged adults are worried about illness, and young adults—both men and women—are worried about unemployment and poverty. The analysis suggests that the Turkana of Lorugum and Kakuma have more worries about jobs, unemployment, education, and school fees than those in Lorengo and Lokichoggio—that is, worries that only emerge when such opportunities exist in an area. The presence of development in both areas might explain why the residents of Lorugum and Kakuma share these worries with other groups across Kenya. At the same time, worries over illness and general health, thirst, and livestock illness, largely held by older men and women across the four locations, disproportionately affect younger and middle-aged men and women in Lorengo and Lokichoggio. The distribution and intensities of worries in Kakuma and Lorugum can be explained by the presence of refugees and relief or development investment, respectively; and in Lorengo, they can be explained by the absence of any development efforts. But Lokichoggio presents an interesting case. As the headquarters for the UN Sudan War efforts from 1988–2006, Lokichoggio had an impressive public service infrastructure for health, education, commerce, and other activities that are central to development. However, after the UN headquarters moved to Juba between 2006 and 2007, Lokichoggio underwent a drastic attrition of jobs among the Turkana and other Kenyan staff unable to move to South Sudan. While the non-Turkana returned to find jobs in other areas of Kenya, the Turkana were faced with the choice of staying in Lokichoggio amid the dwindling resources and population. One man said: “The economy of this place is completely controlled by refugees. The UNHCR is here for the refugees and also most of the NGOs [nongovernmental orga- nizations]. Even the Kenyan government cares more for the refugees. Only workers who stay here are teachers, most other officers come from ‘down- Kenya’ and even they are being supplemented by UNHCR. Look at Loki [choggio], it was so busy, and then when the UNHCR left, it has become a White Elephant, so much building, and now nothing.” (KII 2015). The story of Lokichoggio is a cautionary tale about development based on a single or central activity. When the UN mission, which provided the bulk of the opportunities and capital, left for Juba, Lokichoggio experienced decline within three years, achieving the low levels of nutritional and psychosocial well-being that characterize small and isolated rural settlements like Lorengo. One woman said: 62 Refugee Impacts on Turkana Hosts: A Social Impact Analysis “If the devolution really gives money to the county and the county uses the money well, that would be good for Turkana. And for the educated and non- educated Turkana alike, there will be jobs from the oil project and the water. Like it was at Lokichoggio. When the UN mission was there, Loki was a good place, with jobs and growth. So it can be good, as long as those things, like devolution, like the Oil companies, they stay and treat the Turkana as part- ners, they see them as the people who own the land but are generously shar- ing so should be cared for.” (KII 2015) The concern is that development at Kakuma and Lorugum are, like Lokichoggio, depen- dent on a narrow range of activities; for Kakuma, the refugee camp, and for Lorugum, the hydroelectric power plant. A difference between the two areas is that the residents of Lorugum were able to diversify their subsistence activities, moving from only pas- toralism to include agriculture and other small-scale industry and manufacturing. The proposed development efforts at Kakuma should include investments in diversification, moving away from a reliance on a single activity to provide jobs and opportunities. Gender-Differentiated 6 Norms, Dynamics, and Constraints at Kakuma Oral histories and archaeological evidence suggest that the Turkana once depended heavily on food resources obtained through agriculture and foraging—activities now considered the domain of women and girls. Hence, although the activities of men as pastoralists and women as agriculturalists and foragers seem to follow generally observed gendered dichotomies (Rosaldo 1974), the activities of women were exalted in myth and function during precolonial periods (Lamphear 1988, 1992). However, changes introduced by the British and Kenyan administrations from the 1900s to the 1980s resulted in significant systemic changes to this social organization. Women’s labor shifted from its central role in the subsistence economy and in myth to being relegated to a secondary role within the emergent awi-adakar system. Ironically, while women continued practicing agriculture and foraging and taught younger generations the skills related to food extraction, processing, and storage, as well as the care of young ani- mals; their role and status within the larger society diminished. Pastoralism expanded its significance beyond that of status and prestige, becoming the primary subsistence practice associated with the Turkana. From 1930 to 1980, this change was paralleled by the forced settlement of the Turkana into small settlements that became the boundaries of the domestic realm for women. Interestingly, prevalent narratives from informants confirm various studies on the changes in the lifestyle of the Turkana, suggesting that the discrimination against women and gendered inequities recorded as part of traditional Turkana culture by external observers and development policy makers might in fact have been a response to colo- nialism and more recent changes during the postcolonial period, such as the culture of raiding (ngokoro and moranism) (Hendrickson et al. 1996; Oba 1992). The narrative further suggests that their forced settlement, the downgrading of women’s labor, and the restrictions of women’s activities to the domestic realm may have been internalized by the Turkana as an adaptation to the changing sociopolitical system, intensifying the investment in pastoralism as the primary subsistence activity and the continuation of pastoralism as the primary culturally acceptable means of achieving status and prestige. 63 64 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Mothers and daughters, bakers and entrepreneurs 6. Gender-Differentiated Norms, Dynamics, and Constraints at Kakuma 65 This also disincentivized investments in agriculture and foraging beyond subsistence levels. These social changes were highly correlated with alterations in the economic organization of Turkana County, particularly the decline of status-providing opportu- nities for men and of agency and socioeconomic participation among women. Gendered Vulnerability and Structural Violence among Turkana Men of Kakuma Over the past 20 years, disparities between the development of Turkana and the rest of the country led to an upsurge of external development activities, which have increased dependency but which also led to a decline in pastoralism and an increasing reliance on women’s activities (albeit without status), such as low-energy agriculture (horticulture), foraging, and operating small businesses. The men, however, view pastoralism as their only viable activity, and as they experience its decline, their alternatives are unemploy- ment or menial jobs for development agencies or other groups. Turkana men are often relegated to menial jobs, such as fetching water and food for refugees, digging ditches, working in construction, or providing security for relief and development agencies under the general “naenda Kenya” narrative—a reinforced sense that they are outsiders in their own country and that refugees get better services and resources than they do. A lack of permanent jobs available to Turkana men within Tur- kana—other than unskilled manual ones or pastoralism—also compounds the larger issue—that many of these menial jobs have the potential for interactions of indifference or mistreatment that create experiences and reinforce narratives of indignity (matharau in KiSwahili and ng’imeny in Turkana). The cycle is further reinforced when groups of Turkana men are seen sitting by the roadside in the towns. The emergent narra- tives among external actors—relief and development workers as well as refugees in Kakuma—is of Turkana men being “lazy” and generally “unable to be productive.” One refugee man from the Oromo community explained both sides of the narrative: “When you see a Turkana man like that, sitting there, not doing anything, on his stool, holding his stick, you think, why doesn’t he work? But you know from looking at him that he has lost his herd. I am from a pastoralist com- munity myself. I know what he’s feeling. I lost my herds when the government accused me of being part of the OLF [Oromo Liberation Front]. I could not do anything. I felt that as an Oromo man, I lost myself. That man, there, he is broken. He is broken because when he lost his herd, his animals, he lost him- self. He may start drinking, start abusing drugs, get into fights, but the biggest problem is that he lost his herd, and he lost his pride, his status.” (KII 2010) 66 Refugee Impacts on Turkana Hosts: A Social Impact Analysis This quote is reinforced by other data from informants as well as by studies on the problems of men and the construction of masculinity in marginalized communities where being a “man” is increasingly determined by narrow factors—in this case, the ability to build and maintain large herds and for younger men, to participate in raiding (ngoroko or moran). By comparison, women who were interviewed reported engaging in a diverse range of livelihoods, including agriculture; foraging for wild plant food, such as edapal, elamach, edu, and acacia; making and selling charcoal; collecting and selling firewood; gathering thornbushes and poles for fence-making and house con- struction; constructing fences and walls in the camp; and working menial cleaning jobs in the homes of refugee as well as in the development and relief workers’ compounds. However, engagement with the refuge economy has become increasingly critical for household income, and women are often the central interlocutors—economically and socially—between the refugee and host populations. Turkana women engage refugees in trade for milk, firewood, and charcoal. They also perform a range of domestic tasks, including cleaning, washing clothes, cooking, mending fences, and fetching water. While these activities are crucial for earning an income for the home, they also facilitate regular interactions with refugees, sometimes leading to increased and lasting personal bonds. Many women have used the shops in the camp to buy goods to sell at retail prices in the settlements outside the camp; some have built small businesses selling tobacco, relief food, and other consumables to refugees and the Turkana alike. One particu- lar study explores the distinctive ability of women to foster relationships between the Turkana and refugees (Ohta 2005), noting that these relationships and networks also often lead to increased support from refugees in times of need, especially when Turkana women form friendships with refugee women, attend church and other social functions together, and are subsequently given shelter—and even food—in times of need. For Turkana women in particular, bearing increasing responsibility for providing for the home by engaging in income-generating activities as a matter of survival and sup- port—however marginal—creates a space for their empowerment and agency. At the same time, many of the Turkana—men and women alike—note the risk of assault when asking for payment or attempting to negotiate fees for work. Jobs such as domestic ser- vice in the camp, charcoal and firewood production and sale, and fence-making expose women to the risk of physical and sexual violence. And some women from the refugee and host communities engage in commercial sex work, the cause and effect of which is relative vulnerability. Turkana men have a more varied set of interactions with the refugees in that they ben- efit from providing labor to them and to nongovernmental organizations (NGOs) as construction workers and security guards, respectively. However, their primary services 6. Gender-Differentiated Norms, Dynamics, and Constraints at Kakuma 67 of providing goods, such as large quantities of livestock and charcoal, are controlled by the nimuchurus, the brokers who prevent easy access to the men. The Turkana men are more likely to engage in social interactions, with the refugees consuming the stimulant miraa as well as locally produced alcohol, often resulting in violence. They are also more likely than women to produce the narratives of the refugee as the violent “other.” A significantly important finding in this study and a potential focus for further research is that when the health data were analyzed using regression and controlling for worries (total, economic, and health), the earlier differences in sum of skinfolds (SSF) and body fat percentage (BFP) between Kakuma/Lorugum and Lorengo/Lokichoggio remained significant for the women (appendix D.1), but were significantly lessened for the men of Kakuma (appendix D.2), suggesting that the Turkana men of Kakuma, specifically older and middle-aged men who report greater negative interactions with refugees, might be experiencing a more profound toll on their psychosocial and nutritional well-being than Turkana women. Gender-Based Violence and Structural Violence among the Turkana Women of Kakuma Vulnerability acts as a multiplier for compounding gender-based violence, including sexual and structural violence in the host community of Kakuma, leaving women in liminal positions of social stigmatization and exclusion that affect them as well as their families. According to women interviewed from both the refugee and host communi- ties, the primary causes for both types of violence are discrimination on the basis of ethnicity, language, and religion, manifested in health status and access to cash. The women revealed that in the patriarchal system found among most groups in the refugee and host communities, motherhood—usually seen as a status-enhancing factor—can become a conduit for stigmatization and exclusion among single mothers and their children. Corruption, graft, and discrimination experienced at the hands of develop- ment and relief workers—a cause for concern—are among other factors mentioned. Motherhood can also increase the risk of victimhood and/or exploitation because women with children must bring cash or food home on a daily basis and therefore might be less likely to report assaults or discrimination. An important and recurring issue among male and female Turkana informants at Kakuma was the treatment of women in the awi after the death of their spouses. The majority of informants who reported the death of their fathers or husbands said that they, along with their children, were excluded from the awi and often given minimal control over the livestock in the herds. The interviews revealed a systemic gap in the legal protection given to women in the case of divorce or death of a spouse and the way 68 Refugee Impacts on Turkana Hosts: A Social Impact Analysis women are actually treated. These issues become more dire in the context of a polyga- mous household where arguments between co-wives can cause friction and factional- ism within the awi and even within the adakar. Ethnographic data confirm that the issue of widowhood mentioned in the worry analy- sis is particularly dire for older women because in the event of the death of their spouse, they are left without any protection. Many single mothers either seek the protection of their husband’s brother and become a concubine or co-wife in the brother’s awi or, in the case of divorce or abuse, return to their father’s home, in which case the father’s awi might be asked to pay back the marriage dowry. Because life in the rural settlement is centered around the awi, and because it is the job of the woman to care for the shoats of the awi’s herds, single or abandoned women and their children often resort to moving into areas with more job opportunities, such as Kakuma, Lodwar, and Lokichoggio, and becoming part of the large low-income labor force. Many male and female informants stated that when they moved with their mothers, they also had to participate in those labor activities to contribute to the family’s income. However, many of the informants working at the camp had spouses who also worked there, and/or were unemployed/ herdless and perceived working for the refugees as a contribution to or the primary source of household income. Interestingly, contributors to household income also included children as young as 8–10 years old working in the homes of refugees. The lack of legally enforceable protection for Turkana women affects single women because they can become targets of older men wanting them to become concubines, girlfriends, or co-wives. These arrangements are often financial agreements that involve the man providing a minimal allowance for the woman, predicated on the assumption that the woman would also work to supplement it. This supplementary income can come from a variety of jobs, including collecting and selling firewood and charcoal or working as a domestic servant in the homes of refugees or NGO offices. The larger issue of systemic violence remains a considerable problem. Violence perpetrated against Tur- kana women extends from a number of factors as noted below. n Structural realms. – In a patriarchal and patrilineal system, the social value of a woman is lower than that of a man beginning at birth; – Despite the diversification of labor and despite becoming sole or primary income- generators for families, the labor of women is undervalued compared with that of men; 6. Gender-Differentiated Norms, Dynamics, and Constraints at Kakuma 69 – Diversification of subsistence and labor pathways generally exacts costs in terms of time, health, and emotional and physical well-being; and – Interactions with health, relief, and development workers are marked by experi- ences of exclusion and marginalization. n Physical realms. – The combination of poverty and necessity frequently drive women to seek liveli- hood opportunities at high personal risk, leaving them vulnerable to exploita- tion, robbery, and assault while they collect firewood or charcoal, work in the fields, walk near the refugee camp, or withdraw money from banks and other institutions; – The correlation between thwarted masculinity with perceived low status stem- ming from unemployment or loss of herds and male drinking or other negative coping behaviors is commonly linked with intimate partner and domestic vio- lence, usually of women and children; and – Quarrels between refugees and the Turkana can explode into public, physical, and emotional abuse of Turkana women and children, within and outside of the camp. The complex nature of the changes to the social, economic, and educational/knowledge systems of the Turkana are compounded by a lack of protection for the legal status of women in long-term relationships. Many of our informants noted that one strategy men use to control women’s status and income is the threat of taking another wife. Living under a consistent threat of men acquiring concubines, girlfriends, or co-wives, and seeking to maintain their status within the household, many Turkana women say that this threat serves as a subtle coercion to stay in abusive relationships. They also reported that upon the death or divorce from a spouse, their dead partner’s relatives can dispossess them of any assets. Many respondents said that women facing the death of or abandonment by a spouse find their way to larger labor markets where they can find low-paying jobs and often develop a suite of diverse livelihood activities to help support their families, either by supplementing the family income or often by being the family’s primary provider. An analysis of the data suggests that the structural dynamics of gender among the Turkana of Kakuma are highly complex. They are rooted in traditional gender roles and mores, which are actually relatively recent but which are constantly reinforced, 70 Refugee Impacts on Turkana Hosts: A Social Impact Analysis sometimes by proposed solutions such as access to education—which requires cash— thus reinforcing investments in pastoralism. Investments in education that did not lead to employment left large groups of herdless men vulnerable to depression, disaffection, and a range of negative coping behaviors, such as substance abuse, and has left women vulnerable to gender-based violence and other forms of exploitation and exclusion. In the absence of alternative but culturally appropriate status-generating subsistence path- ways for men, the adaptive strategy has been to reinvest in familiar knowledge systems and production methods that can be controlled through the knowledge. The decline of men in this “lost generation” places the burden of provisioning for the family on women who continue to diversify their subsistence pathways with hard menial labor in exchange for low wages and with no resulting gain in their status. This dynamic process can be summarized as follows: n Women’s lower social status in the awi-adakar system leaves them vulnerable to abandonment or displacement. – They frequently seek out areas such as Kakuma, which have ample low-skill labor opportunities. – They develop multitasking and subsistence diversification activities specifically related to refugees, including collecting firewood, making charcoal, building fences, cleaning houses, washing dishes and clothes, and fetching food and water. – They earn K Sh 100–150/day (US$1–1.50) and frequently food as well for their activities in the refugee camp, which in turn supports their agricultural labor efforts (growing sorghum and vegetables). n Turkana’s recent history suggests that the correlation of men’s status with herd size and the idea that pastoralism is the only significant and legitimate activity for Tur- kana men leave them vulnerable to unemployment, exclusion, and substance abuse. – Men may try to find small jobs, such as in construction or as security guards, but with a view toward building their herds. – Many jobs are hard to get and usually involve some level of graft or connec- tions, and hence those without capital or connections may not find any means of subsistence. – According to this study, among many Turkana men, the lack of a herd and/or job is linked to depression, substance abuse, and lower health status. 6. Gender-Differentiated Norms, Dynamics, and Constraints at Kakuma 71 Although recent efforts in education outreach have transformed the attitudes of many men that pastoralism is the only legitimate means of gaining status, this study dem- onstrates that maintaining herds at Kakuma within a household unit or as part of the awi remains the main pathway to marital alliances (dowry/bridewealth); contingency investment; and quick access cash for medical, educational, or other socioeconomic needs. Ironically, this reinforces the Turkana’s insistence on continuing to engage in pastoralism, which, in turn, reinforces gendered manifestations of structural violence: Turkana women are able to find opportunities and end up multitasking to feed their children while, at the same time, Turkana men feel hampered by their need for herds and might therefore be relegated to menial labor or, in the absence of even that, to depression and substance abuse. Devolution, Oil, and Water: 7 Perspectives and Narratives on Underdevelopment, Recent Discoveries, and Structural Changes Ethnographic and focus group interviews with members of the Turkana community in Lodwar, Lokichoggio, Lorengo, and Lorugum suggest that the Turkana are aware of inequities between themselves and refugees, but that they have generated a social causation narrative in which the primary cause of underdevelopment is systemic cor- ruption in the local, regional, and national political economies, where they are sub- jected to bribery, graft, neglect, indifference, stereotyping, and discrimination. New development efforts are understood through and subsumed by the narrative that any potential benefits from recent changes such as devolution and the discovery of vast oil and water resources are fully dependent on and are necessarily limited by low levels of transparency and accountability and the differential interest shared by the rest of Kenya and indeed the world regarding the resources on Turkana lands as opposed to the well- being of the Turkana people. Even among relief and development workers at Kakuma and other areas in Turkana, there is a distinct and well-maintained hierarchy among Kenyan staff (“down-Kenyans”) and the Turkana, with Turkana staff working in offices and doing the bulk of the menial tasks occupying different social spaces than non-Turkana Kenyans. While overt discrimina- tion against the Turkana is less obvious than the systemic and structural discrimination they face, narratives of the backwardness of the Turkana are common among down- Kenyans. This larger narrative of intransigence and violence in mainstream Kenyan society ignores or dismisses various crucial factors underlying the present decline of the Turkana. As noted earlier, these historical and structural processes have served to generate a consistent narrative of indignity (ng’imeny), which emerges when outsiders come to Turkana for business, employment, development, or to take up residence, or when the Turkana go to other parts of Kenya—as they say, “wakati tumekwenda Kenya” 73 74 Refugee Impacts on Turkana Hosts: A Social Impact Analysis (when we go to Kenya). In turn, the broader narratives of distrust, anger, and exclusion that the Turkana have generated about the political, economic, and social system that has marginalized them for over than a century shapes their perceptions and beliefs regarding any external development interventions. Narratives and Responses to Devolution, Oil, and Water There have been three major economic and sociopolitical developments that have affected Turkana County since 2009: the discovery of oil (2009–12); the discovery of water resources (2013); and the shift toward decentralized devolution across Kenya (2010–13), including greater political and economic investments in Turkana (2010– present). These developments are described briefly in this section and assessed regard- ing Turkana observations, beliefs, and perceptions of the projects’ projected benefits. Beginning in 2006, following the discovery of oil in Uganda, Mozambique, and South Sudan in the early to mid-2000s, the Kenyan government invested in oil and gas explo- ration. In 2012 and 2013, the British firm Tullow Oil announced the discovery of oil in the Lokichar areas of South Turkana County (Blocks 10BB and 13T) in sufficient quantities to warrant further investigation for commercial extraction. Since then, areas 11A and 11B and 10A in northern Turkana are being examined because of their geo- logical similarity to the oil-bearing rocks of South Turkana County. It is hoped that the ongoing extraction efforts will create enough jobs for massive local employment among the Turkana at all levels, from management and technical staff to daily wage laborers. Job growth and an increase of capital and spending based on the new jobs in Turkana are expected to lead to the development of the service and secondary sectors, including trade, business, health, education, and other livelihood opportunities. In September 2013, in partnership with the Kenyan government and the United Nations Educational, Scientific and Cultural Organization (UNESCO), the firm Radar Technologies announced the discovery of a vast underground aquifer, now called the Lotikipi Basin Aquifer. This is the largest aquifer ever discovered in Kenya, possess- ing an estimated 200 billion cubic meters of fresh water and covering an area of 4,164 square kilometers. According to estimates, if properly managed, the water would pro- vide fresh water to the people of Kenya for over 70 years. In an arid region like Turkana, this is especially welcome news, and there are currently a number of plans to extract the water for human use as well as for irrigation and livestock needs. The broad idea is to use the water to introduce extensive and intensive agriculture in Turkana. How- ever, an unverified study by UNESCO (see Avery 2014: 4-5) suggests that high salinity might make the water unfit for human consumption. The report was widely publicized in Turkana, generating public outrage and protests and bolstering ongoing narratives 7. Devolution, Oil, and Water: Perspectives and Narratives on 75 Underdevelopment, Recent Discoveries, and Structural Changes about the purposeful exclusion of the Turkana from Map 7.1. Oil (Red) and Water (Blue) Discoveries resources, even in their own backyard (map 7.1). in Turkana County The Kenya Devolution Act of 2013 marked a formal change in the political system of Kenya, following procedures agreed upon in the new Kenyan con- stitution established in 2010, which pushed for increased accountability and autonomy with regard to resource spending and budgetary decisions for local and regional areas. Policy makers view this as one of the most significant and positive changes in the Kenyan governance processes since indepen- dence because it could give more power and voice to local opinions, needs, and perspectives, espe- cially in the northern marginalized counties such as Turkana, Wajir, and Mandera. In principle, these projects and developments are aimed at making significant and positive changes for the people of Turkana. Respondents were asked to reflect on the role of devolution and the discovery and extraction of oil and water in their lives, at present or in the future. Source: Based on Avery 2014. As expected from previous studies of marginalized groups, the majority of interviewed respondents among the Turkana (n=79) [Kakuma (n=55), Lokichoggio (n=12), Lorugum (n=3), and Lodwar (n=9)] have views that fall within the aforementioned narrative of distrust of and suspicion that the government (local and national) and external actors would downplay any major benefits from the three developments. Some respondents (n=12) initially suggested that devolution in particular, but also the discovery of oil and water, might be good for the Turkana in strikingly similar ways to the eagerness with which some young men in all four areas where interviews were conducted welcomed the news that a refugee camp would be opening in Kakuma in 1993 (as reported by Ohta 2005): Kakuma (n=5), Lorugum (n=3), Lokichoggio (n=3), and Lodwar (n=1). The disparity between Kakuma, Lodwar, and Lokichoggio, where only 10–15 percent of responses are positive, and Lorugum, where almost 50 percent were, can be explained by the greater development benefits ushered in by the Kenyan government, which the people of Lorugum have experienced over the past 20 years. While there is a variation 76 Refugee Impacts on Turkana Hosts: A Social Impact Analysis in the responses to the potential benefits of devolution and oil and water discover- ies, negativity arising from anger and frustration over years of being marginalized by the national government and corruption within the local government, are revealed in almost every interview conducted in Kakuma, Lokichoggio, Lodwar, and Lorugum. Feedback between Emergent and Previous Narratives The narratives of distrust about the potential benefits of devolution, oil, and water are compounded by narratives that blame not only the government systems but also down-Kenyans for taking opportunities and jobs that “should” go to the Turkana and for generating the mainstream narratives of Turkana backwardness and intransigence. However, some respondents differentiate between the Kenyan government, whom they accuse of neglectful policies, and the local government and politicians, whom they accuse of more harmful intentions linked to corruption that directly affect the people. The down-Kenyans who suggest that there will be more jobs for the people of Turkana when the oil and water extraction systems are built also express a note of caution and voice a narrative suggesting that the benefits to the Turkana will be limited. In many cases, the Turkana note that while employment in oil and water extraction is desirable as advertised, it is not easily accessible or beneficial to local residents. The few nationals in management are mainly surveyors, advisors, and engineers. “Skilled” employees are usually nationals, including welders, plumbers, electricians, crane opera- tors, assistant drillers, masons, group financial controllers, ecologists, and sociolo- gists—all mainly from other parts of Kenya. Unskilled workers are all nationals and are predominantly from the local community. Two years ago, hope and enthusiasm were evident at meetings regarding how the oil would be drilled; how it would be shared; and what other benefits it would offer to the local population, such as education and jobs. But the promised benefits have yet to accrue to the Turkana. While Tullow Oil, the agency contracted for oil exploration, is in place, there is real skepticism regarding the terms of the contract, including who has and who will benefit from the deals. For example, Tullow Oil provided education scholarships to the Turkana people that were channeled through political leaders and inequitably distributed. A related issue is the problem of education, presented as the primary pathway for success and encouraged by political and social leaders as well as development actors. However, there are numerous structural problems in the “education as panacea” argument, includ- ing the abuse of children; a lack of money for textbooks, uniforms, and examination fees; and the tremendous disparity between the number of primary versus secondary 7. Devolution, Oil, and Water: Perspectives and Narratives on 77 Underdevelopment, Recent Discoveries, and Structural Changes and tertiary educational institutions. These narratives of suspicion and anger against the political system and the corruption inherent in daily encounters are paralleled by nar- ratives of frustration over the presence of the refugees, although they are balanced by frequent acknowledgment of benefits received due to their presence. One man said: “In many ways we have literally become the clients and the servants of the refugees, and they are the patrons, they are the employers. Without them we, the local Turkana, would not be able to survive. Working as domestic ser- vants, to fetch water, to help them carry food, to be security and watchmen for the NGOs. Can’t we at least once [work] to get a Turkana in a good job?” (KII 2015) Anger aimed at the refugees is mirrored by perceptions that the UNHCR and other external actors have been ineffective at making any real difference. Many respondents suggest that the efforts of the UN, nongovernmental organizations (NGOs), and other external actors were often stymied by the local leadership and political system. A final narrative emerging from the oil and water discoveries is that of theft and rob- bery. Turkana respondents believe that not only would they fail to reap the incidental benefits of oil and water, they would not even have access to those commodities and resources. The Turkana coexist in an uneasy system of exchange with the refugees of Kakuma and with various external development and relief organizations across Tur- kana County, including the UN. However, these narratives of distrust and suspicion are paralleled by narratives of the Turkana receiving benefits and help from the presence of external actors and refugees. Narrative Summary The dominant narratives among the Turkana that emerged in areas such as Lokichog- gio in the northwest, Lorugum in the southwest, Lorengo and Lodwar to the south of Kakuma, and Kakuma are of distrust and anger at the political, economic, and social system of exclusion, marginalization, corruption, and discrimination. n The central administration of the Kenyan government is accused of neglect stretch- ing over 50 years, including deep memories of policies first initiated by the British. n The larger Kenyan society—the “down-Kenyans”—treat the Turkana with disdain at worst and indifference at best, only showing interest in their natural resources. 78 Refugee Impacts on Turkana Hosts: A Social Impact Analysis n General stereotypes about the Turkana combined with nepotism among down- Kenyans reserve the good jobs for the non-Turkana, relegating the Turkana to menial jobs and casual labor. n The local political landscape is characterized by gross corruption and nepotism while the Turkana kinship exchange system demands that any progress made by an individual be shared across the larger extended kin group, including but not limited to help with money, food, jobs, and other benefits within the power of the individual family member to provide. n Educated and uneducated people alike fear being excluded directly by the down- Kenyans from opportunities arising from development or indirectly due to their lack of networks with local elites or capital to pay bribes to prospective hiring personnel. n These narratives generate a hierarchy of suspicion, distrust, and anger. Local politi- cians are held the most responsible for the lack of development and for the poverty; they are followed by down-Kenyans for excluding and discriminating against the Turkana; the government of Kenya for its indifference and neglect toward them; and finally, the UN, development organizations, and the refugees. Daily frustrations further reinforce this larger, complex narrative that seeks and finds multiple targets for the marginalization and impoverishment of the Turkana commu- nity. It is reassuring, however, that the Turkana do not see the refugees as a target but rather understand that the refugees are also in marginalized positions. Modeling the Pathways 8 of Interactions and the Complex Impacts of Refugees on Host Communities A qualitative engineering model has been attempted to examine, understand, and describe how narrative, interaction, and external factors between and within refugee and host communities affect and shape each other. Specifically, the model showcases the complex and dynamic pathway of coexistence in which general and specific narratives held by the Turkana and the refugees shape and are shaped by interactions between the groups and how the narratives of coexistence interface with external factors, such as the climate/environment and drought/famine that generate their own narratives of suffering, resilience, and dignity. Specifically, the analysis shows how the refugee camp and the presence and activities of the refugees have complex effects on the Turkana host community. These effects range from negative impacts that generate acts of violence or indignity to mostly positive impacts, including enhanced social interactions and networks; better access to markets; and higher indicators of health and physical well- being compared with other sites in Turkana, particularly Lorengo and Lokichoggio. The health and physical indicators of the Turkana of Kakuma are similar to those of Lorugum, an area that exhibits more intense development investment. Development investment levels in Kakuma are not as high as in Lorugum. Develop- ment gaps in Kakuma—also seen in Lorengo and Lokichoggio—are filled: n Partially by the presence of relief and development agencies in Kakuma, including access to health care; and n Mainly by the presence of refugees resulting in enhanced access to food, cash, and support generated from daily social and economic interactions between refugees and their host community. 79 80 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Social and economic interactions are not distinct categories: economic transactions are embedded within social interactions, and the resulting experiences further shape the other, positively and negatively. Refugees and the Turkana of Kakuma generate narra- tives about one other as the violent “other”; as difficult, greedy, and irrational people with unreasonable demands for goods and resources to which they are not entitled; and as abusers of women and children. Both groups also generate counternarratives in which they see the other as good people, fellow sufferers, friends, and neighbors, and in which they seek to understand context, contingency, justice, and fairness. Actual interactions shape the perceptions of members of the host community living in and near Kakuma toward refugees in ways that are radically and significantly different than members living elsewhere. The emergent experiences of the interactions between indi- viduals then reinforce, modify, and/or reshape the subnarratives and subsequently, the metanarrative of the refugee as violent “other” or foreign usurper. The metanarrative of the refugee as violent “other” or foreign usurper exists throughout Turkana County, and although the respondents from all four sites in this study share this dominant metanarrative, the Turkana of Kakuma and Lorengo (25 kilometers from Kakuma) who actually interact with the refugees on a daily basis share positive per- ceptions of the refugees—refugees are good and refugees bring benefits—and were far more specific in explaining their narratives. Respondents from Lorugum and Lodwar described the dominant metanarrative without nuance, with generic statements such as “refugees kill the Turkana” and “refugees steal from the Turkana.” The Turkana of Lokichoggio have experienced a decline in their living standards and their well-being over the last 8–9 years since the UNHCR Sudan Relief operations moved its operations from Lokichoggio to Juba, followed by the associ- Figure 8.1. Trends in Positive and Negative ated NGOs and civic bodies. Respondents from Perceptions of Refugees among the Turkana of this area acknowledged that refugees brought Kakuma, Lorengo, Lokichoggio, Lodwar, and Lorugum benefits, but still maintained the dominant meta- 60 Refugees are good narrative of the refugee as the violent “other” or Refugees are bad Percentage of respondents 50 foreign usurper. The worry-perception analysis 40 clearly reveals that the generic negative percep- 30 tions of “refugee as bad” and “refugee as violent 20 other or foreign usurper” are roughly equal across all sites, including Lodwar, where 4 out of 10 10 informants conveyed negative perceptions about 0 0–10 20–30 90–100 120–130 150–160 the refugees (see figure 8.1). (n =150) (n =150) (n =150) (n=10) (n=150) Kakuma Lorengo Lokichoggio Lodwar Lorugum Distance from Kakuma/km However, the analysis also shows that the speci- Source: World Bank survey 2015. ficity and frequency of subnarrative 2—“refugees 8. Modeling the Pathways of Interactions and 81 the Complex Impacts of Refugees on Host Communities are good” and subnarrative 3—“refugees are beneficial” among respondents is inversely proportional to the distance of their homes to Kakuma Refugee Camp. This includes ethnographic data from Lodwar, where two out of 10 informants reported positive per- ceptions of refugees. The data here seems to suggest that the frequency of interactions between host and refugee, which is directly correlated with distance from the camp, is also correlated with the complexity of perceptions held by the host community. In other words, the likelihood that a member of the host community has negative perceptions of refugees does not significantly vary with distance, but the likelihood that a member of the host community would have positive perceptions of refugees is negatively cor- related with distance from the camp. The findings demonstrate that it is feasible to transform the refugee camp into a sustain- able settlement, as proposed by UNHCR, provided that the host community has a voice in the planning and decision making and are themselves integrated into the new settle- ment—refugees and the Turkana would live as neighbors. There are already Turkana settlements at the outskirts of a number of blocks in the camp, and an entire Turkana village lives alongside refugees in Kakuma 2 in a similar manner but for their status. The first step toward establishing such a transformation is to understand the ways that the various narratives, interactions, and external factors shape each other. The next section presents a qualitative engineering model that highlights these interfaced relationships and provides entry points for interventions. Narrative-Interaction Model for Refugee–Host Relationship As seen in the model in figure 8.2, narratives about refugees are driven by interactions and the experiences emerging from them. Interactions are then shaped by internal- ized narratives, which are modified by individuals as a part of their enculturation and upbringing. As anthropologist Itaru Ohta points out, the narrative of the refugee as violent “other” was being generated even before the camp was built in 1992. As one older man expressed it: “In Sudan, people are killing each other. We will not agree to letting dangerous strang- ers come and settle nearby” (Ohta 2005: 229). However, Ohta also reported counternar- ratives of “refugees as beneficial,” especially among young men who “… brought other news […] from Kakuma. They said that many people would be employed in the construction work for the refugee camp. One of them, who had been employed on road construction work before told me excitedly, ‘We can get a big job. We will be paid 50 Shillings a day.’” (Ohta 2005: 229). 82 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Figure 8.2. Narrative-Interaction Model Metanarrative 2 “Naenda Kenya” Metanarrative 1 Neglect, corruption, Refugees as violent other/ discrimination foreign usurpers Climate/environment Violent or Exchange of Subnarrative 1 fractious products Refugees are bad Narratives of Subnarrative 2 Survivial, resilience, hope Refugees are good Exchange of labor Subnarrative 3 Refugees as beneficial Benign solidarity Droughts/Famines building Health, nutrition Source: Based on survey data (2015). The metanarrative of refugee as violent “other” or foreign usurper and the counternar- ratives of refugee as good and beneficial have been interacting with each other for the past 24 years. The interface between narrative and interaction is clearly an extremely complex process with both affecting the other in a cyclical feedback pathway, as shown figure 8.2. Bringing It Together As previously mentioned, in addition to the somewhat contemporary metanarrative 1 emerging out of Turkana County that refugees are violent “others” or foreign usurp- ers, there is an older, broader, and more dominant metanarrative 2: “naenda Kenya,” which speaks to a sense of general neglect, political corruption, and social discrimina- tion, which began during the British colonial period and persists to this day under the Kenyan government and among mainstream Kenyan society. 8. Modeling the Pathways of Interactions and 83 the Complex Impacts of Refugees on Host Communities Finding 1: History, Experience, Narratives, and Interactions Shape One Other While “naenda Kenya” dominates conversations and generates anger, conflict, and even violence among the Turkana, it is important to remember that similar metanar- ratives exist in most of the northern counties in Kenya, including Wajir, Marsabit, and Mandera. The metanarrative of rampant corruption and marginalization across social, economic, and political forums at local, regional, and national levels is not unique, as similar narratives exist in almost every nation and society in the world. However, this particular metanarrative feeds into fears and concerns among the Turkana that their leaders and down-Kenyans—political and economic elites and mainstream society— would prevent them from benefiting from any developmental interventions and struc- tural changes, such as the extraction of oil and gas in Lokichar, the discovery of water in Lotikipi, or the devolution process. The interactions between the Turkana and their political system, economic infrastructure, social landscape, and external development and relief actors serves to further reinforce and reproduce this narrative and generate an individual and communal sense of indignity (ng’imeny) in which the Turkana view themselves as stuck at the bottom of the Kenyan hierarchy, deliberately excluded from social mobility and well-being. There are, however, other narratives emerging from external factors, such as the cli- mate, the environment, and relief and development efforts, in which the Turkana view themselves as people who endure, as people with a proud history and a glorious past, and as people who know how to survive and outlast some of the worst adversities that their environment and the climate have leveled at them, including receding water levels in Lake Turkana; periodic and frequent droughts, especially in the last 50 years; and famines emerging from many of these drought events. The indignities they suffer dur- ing their interactions with external actors and refugees, including verbal insults, physi- cal beatings, and indifferent treatment, can occasionally explode into violence between individuals, groups, or communities. These incidents usually become internalized and feed back into the subnarratives and metanarratives. Finding 2: Complex Interactions This study demonstrates that indignities are often mitigated by experiences with dig- nity, such as friendship, support, bond-friendships, and generosity shown by refugees and external actors, becoming pathways for the Turkana to maintain their resilience and hope. Women are exposed to physical and sexual violence in their economic 84 Refugee Impacts on Turkana Hosts: A Social Impact Analysis interactions, but they deal with such dangers with incredible courage, overcoming their own fears and bearing great pain as they keep in mind their larger goal of feeding their children, sending them to school, and working toward a better future for them. Policy recommendations emerging from this study that draw on the resilience, dignity, and hope demonstrated by the Turkana and the refugee communities focus on positively changing the nature of interactions and narratives. Finding 3: Refugees Have A Net Positive Impact on Turkana Hosts at Kakuma Returning to the original expectations of this study, results show that while the impacts of the presence of a large number of refugees on the Turkana of Kakuma are both posi- tive and negative, the final analysis demonstrates a net positive impact in terms of food and nutritional intake and quality of interactions. The expected variation is found in the well-being of the host communities of Kakuma and the other study sites—Loru- gum, Lorengo, Lokichoggio, and Lodwar—with overlap between Kakuma, Lodwar, and Lorugum. The study also finds that Lorengo and Lokichoggio have the worst outcomes for social, physical, and mental well-being, potentially due to the absence or removal of Hardware and other sundries in Kakuma 2 8. Modeling the Pathways of Interactions and 85 the Complex Impacts of Refugees on Host Communities development and relief organizations, respectively. The well-being of the host commu- nities in Kakuma follow different pathways than do the host communities at the other study sites due to the impact of both the refugee presence and the daily interactions between the refugees, hosts, and relief and development organizations. The presence of the relief and development infrastructure and refugees is correlated with greater access to jobs, wage labor, and other opportunities for the Turkana of Kakuma, but also with a greater potential for conflict and distrust between the refugees and their hosts. Finding 4: Variations Underlying Positive and Negative Impacts The study notes that despite the greater access to jobs, wages, and development oppor- tunities in Kakuma, there is wide variation in the well-being of Turkana living in Kakuma. This variation was similar to but also greater than that measured in Lorugum and observed ethnographically in Lodwar. It included worries about education, fees, and other services that were not mentioned in Lorengo and Lokichoggio, also sug- gesting that the presence of the refugees and development could be contributing to underlying inequities between and within settlements. This is especially true among the men of Kakuma, who do not fare significantly better in their measures of health than do the men at other sites when controlling for other factors, such as worries. Kakuma men also report more negative interactions with refugees than women do. The ethnographic and survey analyses suggest that the level of development invest- ment is a significant factor in understanding the differential trajectories of well-being across Turkana County where it is directly correlated with social, economic, psycho- social, and physical well-being. Finding 5: Turkana Attitudes, Evaluations, and Hopes A critical concern regarding projected development activities is a confirmation of the study expectations about the perceptions of the Turkana regarding their local and national government and regarding ongoing development initiatives. The vast major- ity of the people interviewed expressed widespread distrust in the ability of their gov- ernment and associated bodies and organizations to deliver the promised benefits of devolution or the oil and water discoveries in Turkana. The study suggests that there is a general disenfranchisement of youth and women in Turkana and a significant gap between political and social elites and other Turkana residents. Structural and physical violence were disproportionately manifested along the categories of gender and age, whereby women are most negatively affected by systemic exclusion from participation in the social, economic, and political spheres. 86 Refugee Impacts on Turkana Hosts: A Social Impact Analysis One of the goals of this study is to develop policy and operational recommendations for changes in structure and action that would benefit both the host and the refugee communities. Respondents suggest a pathway that would help alter the narratives and the interactions. The study reveals how narratives and interactions generate and repro- duce experiences and feelings of indignity (ng’imeny/matharau) or dignity (arimatoi/ heshima) among the Turkana residents of Kakuma, and, going forward, another pair of useful words for shaping narratives and interactions is etic/aking’arakin. Both words are used by the Turkana to describe work or labor, but etic is used to indicate “working for non-Turkana” and aking’arakin is used to indicate “assisting fellow Turkana.” Many respondents explained the difference between the terms, including the implicit hierar- chy in the term etic, even in partnership relationships, and the implicit parity conjured by the term aking’arakin, even when applied to master and servant relationships. The narrative must be altered by changes in attitude from ng’imeny to arimatoi and by trans- forming the nature of interactions from etic to aking’arakin. 9 Conclusion The findings of the study suggest that the narrative of refugee as violent “other” or foreign usurper is nuanced. The Turkana of Kakuma and, to some extent, Lorengo, give specific descriptions of positive perceptions, while those in other areas, such as Lorugum, Lodwar, and Lokichoggio, remain largely negative and general. The findings also demonstrate that the refugees of Kakuma have a significant positive impact on the Turkana of Kakuma due to economic and social interactions that result in greater access to food and nutritional well-being and the presence of relief services that serve the Turkana in addition to the refugees. The counterfactual sites, such as Lorengo and Lokichoggio, present significantly lower indicators of health and physical well-being. The similarities between Kakuma and Lorugum regarding their health data and between Kakuma and Lodwar regarding their ethnographic data suggest that while development services might explain the higher indicators of physical and social well-being in Lodwar and Lorugum, in Kakuma, they can be directly attributed to the presence and activi- ties of the refugees, either indirectly through the UNHCR and other bodies or directly through interactions with the refugees. The Turkana believe that they have been neglected and discriminated against for the 52 years since Kenya gained its independence, and they have struggled with natural and anthropogenic disasters and endemic structural violence stemming from their low position in the Kenyan social, economic, and political hierarchy. Their sense of history has led to a developed narrative in which they do not believe that their own leaders or their fellow citizens in Kenya will help them or let them benefit from any development effort. Even their attempts to enroll their children in educational programs is met with a constant need of cash for uniforms, fees, books, transportation, food, and lodging, required for both private and public schools. Ironically, this has led to reinvestment in pastoralism out of necessity because livestock remain the primary source of immediate cash for most people in Turkana County and the only means of livelihood for those who are unable to attend school. 87 88 Refugee Impacts on Turkana Hosts: A Social Impact Analysis Yet, despite their sense of despair regarding the structure of society and the adversities of life, without exception, the men and women of Turkana who were interviewed for this study had hope for their children’s futures. They have not given up on going into the camp or into the bush, walking in known dangerous zones, seeking food and work, or braving indifference in order to provide food for their families and cash for their pockets. They deal with spouses and family members who suffer from illness, disability, and substance abuse, and many battle these conditions themselves. Many political leaders among the Turkana have been born into and have worked to conquer similar adversities. The people of Turkana who participated in this study want partnerships with their political leaders and want to engage in a socioeconomic and political contract. The recent processes of devolution and the discovery of oil, gas, and water offer the leadership and the community an excellent opportunity to embark on a renewed partnership based on transparency and equity, which could benefit the Tur- kana and decisively change the metanarrative of “naenda Kenya,” which sums up the systemic neglect and marginalization of the Turkana for more than 50 years. Such a partnership would enable the people of Turkana to continue to strive and work in the face of any adversity in order to participate in the general development of Turkana County and their own settlements and cities. The people of Turkana have been exemplary hosts to the refugees of Kakuma and when they live in close proximity have shown them the graciousness to develop nuanced nar- ratives about their goodness and the benefits brought by their presence. In return for this generosity, the Turkana want to engage the refugees and the larger relief process in a partnership to develop Kakuma as a sustainable and thriving settlement, to col- lectively move from etic to aking’arakin and to transition from ng’imeny/matharau to arimatoi/heshima. As the national and county governments make efforts to promote more integrated and sustainable development for the Turkana in partnership with the range of develop- ment and humanitarians agencies supporting the Kakuma refugee camp, the Kalobeyei settlement and the Turkana host community, it is critical to embed a set of key policy options emerging from this study in the planning, investment prioritization, and imple- mentation processes. Acknowledging the value and critical role of community land in the Turkana’s social, cultural, and economic realms, UNHCR and the Turkana County government engaged the Kalobeyei host community in earmarking and allocating land for the Kalobeyei settlement. Inclusive and participatory dialogue and consultations were organized with the Kalobeyei community and the Community Dialogue and Development Committee (CDDC), which represents them, along with the Turkana West opinion leaders, politi- cal leaders, county government executive members, Department of Refugee Affairs, and UNHCR. Mechanisms for the transparent, meaningful, and active participation of communities in the planning, prioritization, and allocation of resources is critical to ensure that the Turkana feel included, empowered, and capable of making key decisions regarding their lives, livelihoods, and resources. Communities have expressed a strong prefer- ence for direct communication with the county administration and for other agencies, including the UN and other international bodies providing developmental support. The community understands the ambitious devolution process to be a well-intentioned and promising institutional change. However, failure to strengthen the presence of the state in Turkana County in a transparent and accountable manner could potentially exacer- bate the existing marginalization, ethnic divisions, local grievances, and conflicts. The increasing emphasis on education among the Turkana is an important opportunity for increased investments by the government on quality education, motivating teachers, and pedagogical skills to address the specific learning needs of the Turkana, along with specific efforts at making education accessible and affordable—both key constraints emphasized by the community. Culturally and socially appropriate options must be explored, considering the predominantly pastoral and dispersed population. The lack of skills and education among Turkana youth could be overcome through on-the-job training schemes focused on the specific skills needed for the new jobs. Tailored pro- grams for enhancing technical along with life skills as well as basic and financial literacy would better equip them to seek employment in emergent areas. The Turkana tend to invest in both the pastoralist economy and in more urban liveli- hoods, each supporting their family in times of crisis, drought, or distress; the ability to access either set of livelihoods serves as an insurance and coping strategy. However, the Turkana’s traditional pastoralist lifestyle is increasingly threatened by various dynam- ics, such as the degradation of grazing land, insecurity, and an increase in population. The national and local government must ensure adequate support for pastoralism—a key survival strategy for the Turkana. Engagement with the economy among women, including those in the refugee camp, has become increasingly critical to household incomes. This participation in income- generating activities creates a space for a woman’s empowerment and agency. However, women participating in income-generating activities within and outside the refugee 89 communities in Kakuma are regularly exposed to physical and sexual violence as well as emotional and psychological abuse. Safe and remunerative livelihoods for women with improved access to input and output markets for agricultural and allied activities as well as the establishment of viable enterprises to augment family incomes, including capital and entrepreneurial training and inputs, must be supported. Opportunities for building on dimensions of local resilience include establishing the presence of the state; consolidating local institutions, such as traditional mediation mechanisms that maximize the role of local leaders among elders, youth, and women; and improving social, economic, and poverty outcomes for the Turkana. An integrated and sustainable development approach is crucial to ensuring steady economic growth, social inclusion, and peace and stability in Turkana. 90 References KII = key informant interview Agier, M. 2002. “Between War and City Towards an Urban Anthropology of Refugee Camps.” Ethnogra- phy 3(3): 317–41. ———. 2011. Managing the Undesirables. Polity. Anderson, D. M., and V. Broch-Due. 2000. The Poor Are Not Us: Poverty and Pastoralism in Eastern Africa. James Currey Ltd. Aukot, E. 2003. “‘It Is Better to Be a Refugee Than a Turkana in Kakuma’: Revisiting the Relationship between Hosts and Refugees in Kenya.” Refuge: Canada’s Journal on Refugees 21(3). Avery, Sean. 2014. “The Turkana Aquifer Discoveries and Development Proposals.” Discussion Brief. DCLI. http://www.fao.org/fileadmin/user_upload/drought/docs/Final_discussion%20paper_Turkana %20aquifer.pdf. Betts, A. 2009. “Development Assistance and Refugees: Towards a North-South Grand Bargain.” Forced Migration Policy Brief 2, Refugee Studies Centre, Oxford Department of International Development, University of Oxford. ———. 2014. “Refugee Economies: Rethinking Popular Assumptions.” The Humanitarian Innovation Project, University of Oxford. Brainard, J. 1986. “Differential Mortality in Turkana Agriculturalists and Pastoralists.” American Journal of Physical Anthropology 70(4): 525–36. Broch-Due, V., and D. L. Hodgson. 2000. “The Fertility of Houses and Herds.” Rethinking Pastoralism in Africa, 165–85. Bush, J. 1995. “The Role of Food Aid in Drought and Recovery: Oxfam’s North Turkana (Kenya) Drought Relief Programme, 1992–94.” Disasters 19(3): 247–59. Campbell, B., P. B. Gray, and P. Leslie. 2005. Age‐related Changes in Body Composition among Turkana Males of Kenya. American Journal of Human Biology 17(5): 601–10. Chkam, H. 2016. “Aid and the Perpetuation of Refugee Camps: The Case of Dadaab in Kenya 1991–2011.” Refugee Survey Quarterly. Doi: 10.1093/rsq/hdw005. Crisp, J. 2003. “No Solution in Sight: the Problem of Protracted Refugee Situations in Africa.” Working Paper, Center for Comparative Immigration Studies. 91 92 References Ebei P., G. Oba, and T. Akuja. 2007. “Long-term Impacts of Droughts on Pastoral Production and Trends in Poverty in North-western Kenya: An Evaluation of 14-year Drought Early Warning Data Series.” In Droughts: Causes, Effects and Predictions, edited by J. M. Sanchez, 103–138. Hauppauge, NY: Nova Science Publishers. Ekaya, W. N. 1998. “Primary Productivity, Energy Flow and Nitrogen Cycling in a Rangeland Ecosystem.” PhD thesis, University of Nairobi. Fratkin, E. 1986. “Stability and Resilience in East African Pastoralism: The Rendille and the Ariaal of Northern Kenya.” Human Ecology 14(3): 269–86. ———. 1997. “Pastoralism: Governance and Development Issues.” Annual Review of Anthropology, 235–61. ———. 2001. “East African Pastoralism in Transition: Maasai, Boran, and Rendille Cases.” African Studies Review 44(3): 1–25. Grabska, K. 2011. “Constructing ‘Modern Gendered Civilised’ Women and Men: Gender-Mainstreaming in Refugee Camps.” Gender and Development 19(1): 81–93. Grawert, E. 2010. After the Comprehensive Peace Agreement in Sudan. Boydell and Brewer. Gulliver, P. H. 1951. “A Preliminary Survey of the Turkana.” In Communications from the School of African Studies 26, Cape Town: University of Cape Town. ———. 1955. The Family Herds: A Study of the Pastoral Tribes in East Africa, the Jie and Turkana. Routledge and Kegan Paul. Harmand, S., and others. 2015. “3.3-million Year Old Stone Tools from Lomekwi 3, West Turkana, Kenya.” Nature 521: 310–15. Hendrickson, D., R. Mearns, and J. Armon. 1996. “Livestock Raiding Among the Pastoral Turkana of Kenya: Redistribution, Predation and the Links to Famine.” IDS Bulletin 27(3): 17–30. Herskovits, M. J. 1926. “The Cattle Complex in East Africa.” American Anthropologist 28(1):230–272. Hilhorst, D., and B. J. Jansen. 2010. “Humanitarian Space as Arena: A Perspective on the Everyday Politics of Aid.” Development and Change 41(6): 1117–39. Hodgson, D. L. 2000. Rethinking Pastoralism in Africa. Suffolk, United Kingdom: James Currey. Horn, R. 2010. “Responses to Intimate Partner Violence in Kakuma Refugee Camp: Refugee Interactions with Agency Systems.” Social Science and Medicine 70(1): 160–68. Humphrey, J. H. 2009. “Child Undernutrition, Tropical Enteropathy, Toilets and Handwashing.” The Lancet 374: 1032–35. References 93 Jacobsen, K. 2002. “Can Refugees Benefit the State? Refugee Resources and African Statebuilding.” The Journal of Modern African Studies 40(4): 577–96. ———. 2005. The Economic Life of Refugees. Kumarian Press. Jones, E. C., and A. D. Murphy, eds. 2009. The Political Economy of Hazards and Disasters. Lanham, MD: Altamira Press, Rowman and Littlefield. KDHS (Kenya Demographic and Health Survey). 2014. Kenya National Bureau of Statistics. https:// dhsprogram.com/pubs/pdf/FR308/FR308.pdf. Lamphear, J. 1988. “The People of the Grey Bull: The Origin and Expansion of the Turkana.” The Journal of African History 29(1): 27–39. ———. 1992. The Scattering Time: Turkana Responses to Colonial Rule. Oxford: Oxford University Press. Mahmud, Iffat, and Nkosinathi Mbuya. 2016. Water, Sanitation, Hygiene, and Nutrition in Bangladesh: Can Building Toilets Affect Children’s Growth? World Bank Studies. Washington, DC: World Bank. doi: 10.1596/978-1-4648-0698-8. Martin, A. 2005. “Environmental Conflict Between Refugee and Host Communities.” Journal of Peace Research 42(3): 329–46. Mburu, N. 2001. “Firearms and Political Power: The Military Decline of the Turkana of Kenya.” Nordic Journal of African Studies 10(2): 148–62. Murphy, D. J. 2014. “Booms and Busts: Asset Dynamics, Disaster, and the Politics of Wealth in Rural Mongolia.” Economic Anthropology 1(1): 104–123. Oba, G. 1992. Ecological Factors in Land Use Conflicts, Land Administration and Food Insecurity in Turkana, Kenya. London: ODI. ———. 2001. “Motivations, Negotiations, and Animal Individuality.” Nilo-Ethiopian Studies 2001(7): 45–61. ———. 2002. “Multiple Socio-economic Relationships Improvised Between the Turkana and Refugees in Kakuma Area, North Western Kenya.” COE International Symposium. ———. 2005. “Coexisting with Cultural ‘Others’: Social Relationships between the Turkana and the Refu- gees at Kakuma, Northwest Kenya.” Pastoralists and Their Neighbors in Asia and Africa 69: 227–39. Oka, R. 2011. “Unlikely Cities In The Desert: The Informal Economy As Causal Agent For Permanent ‘Urban’ Sustainability In Kakuma Refugee Camp, Kenya.” Urban Anthropology and Studies of Cultural Systems and World Economic Development, 223–62. ———. 2014. “Coping with the Refugee Wait: The Role of Consumption, Normalcy, and Dignity in Refu- gee Lives at Kakuma Refugee Camp, Kenya.” American Anthropologist 116(1): 23–37. 94 References Oka, Rahul C., Yang Yang, Ahmed Ali Abdi, and Nitesh Chawla. Under review. “Rapid De-Regulation of Economies is Correlated with Predatory Competition, Cronyism, and Instability.” PLoS One. Oliver-Smith, A. 1996. Anthropological Research on Hazards and Disasters. Annual Review of Anthropology 25: 303–28. Pérouse de Montclos, M., and P. M. Kagwanja. 2000. “Refugee Camps or Cities? The Socio-economic Dynamics of the Dadaab and Kakuma Camps in Northern Kenya.” Journal of Refugee Studies 13(2): 205–22. Pike, I. L. 1999. “Age, Reproductive History, Seasonality, and Maternal Body Composition During Preg- nancy for Nomadic Turkana of Kenya.” American Journal of Human Biology 11(5): 658–72. Pike, I. L. 2004. “The Biosocial Consequences of Life on the Run: A Case Study from Turkana District, Kenya.” Human Organization 63(2): 221–35. Pike, I. L., and S. R. Williams. 2006. “Incorporating Psychosocial Health into Biocultural Models: Prelimi- nary Findings from Turkana Women of Kenya.” American Journal of Human Biology 18(6): 729–40. Porter, G., K. Hampshire, P. Kyei, M. Adjaloo, G. Rapoo, and K. Kilpatrick. 2008. “Linkages Between Livelihood Opportunities and Refugee–Host Relations: Learning from the Experiences of Liberian Camp-based Refugees in Ghana.” Journal of Refugee Studies 21(2): 230–52. Rosaldo, M. Z. 1974. “Woman, Culture, and Society: A Theoretical Overview.” In Woman, Culture and Society, edited by M. Z. Rosaldo, L. Lamphere, and J. Bamberger, 17–42. Stanford University Press. Sanghi, A, S. Ramachandran, A. de la Fuente, M. Tonizzo, S. Sahin, and B. Adam. 2010. Natural Hazards, Unnatural Disasters: The Economics of Effective Prevention. United Nations and the World Bank 2512. http://elibrary.worldbank.org/doi/abs/10.1596/978-0-8213-8050-5. Sanghi, Apurva, Harun Ondar, and Varalakshmi Vemuru. 2016. “Yes” In My Backyard? The Economics of Refugees and their Social Dynamics in Kakuma, Kenya. Washington, DC: World Bank. Simpson, G. L. 1996. “Frontier Banditry and the Colonial Decision-making Process: The East Africa Pro- tectorate’s Northern Borderland Prior to the First World War.” The International Journal of African Historical Studies 29(2): 279–308. Veney, C. R. 2007. Forced Migration in Eastern Africa: Democratization, Structural Adjustment, and Refu- gees. Palgrave Macmillan. World Bank. 2015. World Development Report 2015: Mind, Society, and Behavior. Washington, DC: World Bank. http://www.worldbank.org/en/publication/wdr2015. Wright, D. K., Forman, S. L., Kiura, P., Bloszies, C., and Beyin, A. 2015. “Lakeside View: Sociocultural Responses to Changing Water Levels of Lake Turkana, Kenya.” African Archaeological Review 32(2): 335–67. Appendixes 95 A Mean and Variation of Body Mass Index and Sum of Skinfolds Table A.1. Average Body Mass Index and Sum of Skinfolds for Women Descriptive Statisticsa Standard Number of Location Mean Deviation Observations Body mass index Kakuma Town 19.2873 3.15440 79 Lokichoggio 19.1712 2.17121 72 Lorengo 18.6555 2.12156 65 Lorugum 19.6119 3.17340 77 Total 19.2039 2.74313 293 Sum of skinfolds Kakuma Town 30.850 19.8243 79 Lokichoggio 23.398 12.6817 72 Lorengo 17.572 7.2911 65 Lorugum 33.173 20.6759 77 Total 26.684 17.4341 293 (continued) 97 98 Appendixes Table A.1. Continued Pairwise Comparisonsa Mean 95% Confidence Interval for Dependent Difference Standard Differenceb Variable (I) Location (J) Location (I–J) Error Sig.b Lower Bound Upper Bound Body mass index Kakuma Town Lokichoggio 0.116 0.446 0.795 -0.761 0.994 Lorengo 0.632 0.458 0.169 -0.270 1.534 Lorugum -0.325 0.438 0.460 -1.187 0.534 Lokichoggio Kakuma Town -0.116 0.446 0.795 -0.994 0.761 Lorengo 0.516 0.468 0.272 -0.406 1.437 Lorugum -0.441 0.449 0.327 -1.324 0.442 Lorengo Kakuma Town -0.632 0.458 0.169 -1.534 0.270 Lokichoggio -0.516 0.468 0.272 -1.437 0.406 Lorugum -0.956* 0.461 0.039 -1.864 -0.049 Lorugum Kakuma Town 0.325 0.438 0.460 -0.538 1.187 Lokichoggio 0.441 0.449 0.327 -0.442 1.324 Lorengo 0.956 0.461 0.039 -0.049 1.864 Sum of skinfolds Kakuma Town Lokichoggio 7.452* 2.676 0.006 2.185 12.719 Lorengo 13.279* 2.750 0 7.865 18.692 Lorugum -2.323 2.630 0.378 -7.500 2.854 Lokichoggio Kakuma Town -7.452* 2.676 0.006 -12.719 -2.185 Lorengo 5.827* 2.810 0.039 0.295 11.358 Lorugum -9.775* 2.693 0 -15.075 -4.475 Lorengo Kakuma Town -13.279* 2.750 0 -18.692 -7.865 Lokichoggio -5.827* 2.810 0.039 -11.358 -.295 Lorugum -15.601* 2.767 0 -21.047 -10.156 Lorugum Kakuma Town 2.323 2.630 0.378 -2.854 7.500 Lokichoggio 9.775 2.693 0 4.475 15.075 Lorengo 15.601* 2.767 0 10.156 21.047 a. Sex = female. b. Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjustments) *The mean difference is significant at the level indicated in the table under the column Sig. Source: World Bank survey 2015. A. Mean and Variation of Body Mass Index and Sum of Skinfolds 99 Table A.2. Average Body Mass Index and Sum of Skinfolds for Men Descriptive Statisticsa Standard Number of Location Mean Deviation Observations Body mass index Kakuma Town 18.1896 2.46040 66 Lokichoggio 18.0480 1.71581 74 Lorengo 18.2120 1.65018 74 Lorugum 18.2853 2.30303 71 Total 18.1825 2.03889 285 Sum of skinfolds Kakuma Town 17.121 9.4846 66 Lokichoggio 14.293 3.0235 74 Lorengo 14.096 3.0744 74 Lorugum 16.275 7.5653 71 Total 15.390 6.4122 285 (continued) 100 Appendixes Table A.2. Continued Pairwise Comparisonsa Mean 95% Confidence Interval for Dependent Difference Standard Differenceb Variable (I) Location (J) Location (I–J) Error Sig.b Lower Bound Upper Bound Body mass index Kakuma Town Lokichoggio 0.142 0.347 0.683 -0.541 0.824 Lorengo -0.022 0.347 0.949 -0.705 0.660 Lorugum -0.096 0.350 0.785 -0.785 0.594 Lokichoggio Kakuma Town -0.142 0.347 0.683 -0.824 0.541 Lorengo -0.164 0.337 0.627 -0.827 0.499 Lorugum -0.237 0.340 0.486 -0.907 0.432 Lorengo Kakuma Town 0.022 0.350 0.949 -0.660 0.705 Lokichoggio 0.164 0.340 0.627 -0.499 0.827 Lorugum -0.073 0.340 0.830 -0.743 0.596 Lorugum Kakuma Town 0.096 0.350 0.785 -0.594 0.785 Lokichoggio 0.237 0.340 0.486 -0.432 0.907 Lorengo 0.073 0.340 0.830 -0.596 0.743 Sum of skinfolds Kakuma Town Lokichoggio 2.828* 1.069 0.009 -0.723 4.933 Lorengo 3.024* 1.069 0.005 0.919 5.129 Lorugum 0.846 1.080 0.434 -1.280 2.972 Lokichoggio Kakuma Town -2.828* 1.069 0.009 -4.933 -0.723 Lorengo 0.196 1.038 0.850 -1.848 2.240 Lorugum -1.982 1.049 0.060 -4.047 0.084 Lorengo Kakuma Town -3.024* 1.069 0.005 -5.129 -0.919 Lokichoggio -0.196 1.038 0.850 -2.240 1.848 Lorugum -2.178* 1.049 0.039 -4.244 -0.113 Lorugum Kakuma Town -0.846 1.080 0.434 -2.972 1.280 Lokichoggio 1.982 1.049 0.060 -0.84 4.047 Lorengo -2.178* 1.049 0.039 0.113 4.244 a. Sex = female. b. Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjustments) *The mean difference is significant at the level indicated in the table under the column Sig. Source: World Bank survey 2015. A. Mean and Variation of Body Mass Index and Sum of Skinfolds 101 1. BMI Pastoralist Women—Descriptive Statistics ----------------------------------------------------------------------------------------- -> male = Female Means, Standard Deviations and Frequencies of BMI | refliveli | hood2 Location | 0 | Total -----------+-----------+---------- Kakuma | 15.653333 | 15.653333 | 2.9490732 | 2.9490732 | 3 | 3 -----------+-----------+---------- Lokichogg | 18.951 | 18.951 | 1.7084195 | 1.7084195 | 10 | 10 -----------+-----------+---------- Lorengo | 18.630811 | 18.630811 | 2.0583507 | 2.0583507 | 74 | 74 -----------+-----------+---------- Lorugum | 19.313334 | 19.313334 | 3.0035092 | 3.0035092 | 6 | 6 -----------+-----------+---------- Total | 18.613226 | 18.613226 | 2.1580039 | 2.1580039 | 93 | 93 2. BMI Pastoralist Women—Regression ----------------------------------------------------------------------------------------- -> male = Female Source | SS df MS Number of obs = 93 -------------+---------------------------------- F(4, 88) = 1.69 Model | 30.6199018 4 7.65497546 Prob > F = 0.1587 Residual | 397.82232 88 4.52070818 R-squared = 0.0715 -------------+---------------------------------- Adj R-squared = 0.0293 Total | 428.442221 92 4.65698067 Root MSE = 2.1262 ------------------------------------------------------------------------------ bmi | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | -.0032043 .0141351 -0.23 0.821 -.0312948 .0248863 | loc | Lokichoggio | 3.262099 1.4084 2.32 0.023 .4631999 6.060999 Lorengo | 2.941927 1.261977 2.33 0.022 .4340126 5.449842 Lorugum | 3.579359 1.54496 2.32 0.023 .5090768 6.649642 | _cons | 15.83598 1.468351 10.78 0.000 12.91794 18.75402 ------------------------------------------------------------------------------ 102 Appendixes 3. BMI Pastoralist Men—Descriptive Statistics ----------------------------------------------------------------------------------------- -> male = Male Means, Standard Deviations and Frequencies of BMI | refliveli | hood2 Location | 0 | Total -----------+-----------+---------- Kakuma | 17.99 | 17.99 | 1.6643308 | 1.6643308 | 3 | 3 -----------+-----------+---------- Lokichogg | 18.76579 | 18.76579 | 1.6575363 | 1.6575363 | 19 | 19 -----------+-----------+---------- Lorengo | 18.212027 | 18.212027 | 1.6501822 | 1.6501822 | 74 | 74 -----------+-----------+---------- Lorugum | 19.126363 | 19.126363 | 1.8634871 | 1.8634871 | 11 | 11 -----------+-----------+---------- Total | 18.398131 | 18.398131 | 1.6822335 | 1.6822335 | 107 | 107 . bysort male: reg bmi age i.loc if reflivelihood2==0 4. BMI Pastoralist Men—Regression ----------------------------------------------------------------------------------------- -> male = Male Source | SS df MS Number of obs = 107 -------------+---------------------------------- F(4, 102) = 2.35 Model | 25.2976123 4 6.32440307 Prob > F = 0.0593 Residual | 274.672809 102 2.69287068 R-squared = 0.0843 -------------+---------------------------------- Adj R-squared = 0.0484 Total | 299.970422 106 2.82990964 Root MSE = 1.641 ------------------------------------------------------------------------------ bmi | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | -.0204977 .0090439 -2.27 0.026 -.0384362 -.0025593 | loc | Lokichoggio | .8006022 1.019546 0.79 0.434 -1.221662 2.822866 Lorengo | .5214594 .9754321 0.53 0.594 -1.413305 2.456224 Lorugum | 1.363701 1.073541 1.27 0.207 -.7656628 3.493065 | _cons | 18.68692 .9960794 18.76 0.000 16.7112 20.66264 ------------------------------------------------------------------------------ A. Mean and Variation of Body Mass Index and Sum of Skinfolds 103 5. SSF Pastoralist Women—Descriptive Statistics bysort male: tab loc reflivelihood2 if reflivelihood2==0, sum(skinsum) ----------------------------------------------------------------------------------------- -> male = Female Means, Standard Deviations and Frequencies of Sum of Skinfold Thickness (mm) | refliveli | hood2 Location | 0 | Total -----------+-----------+---------- Kakuma | 15.733334 | 15.733334 | 3.8475103 | 3.8475103 | 3 | 3 -----------+-----------+---------- Lokichogg | 33.5125 | 33.5125 | 26.507704 | 26.507704 | 8 | 8 -----------+-----------+---------- Lorengo | 17.570769 | 17.570769 | 7.2933344 | 7.2933344 | 65 | 65 -----------+-----------+---------- Lorugum | 45.166667 | 45.166667 | 22.387021 | 22.387021 | 6 | 6 -----------+-----------+---------- Total | 21.078049 | 21.078049 | 14.256123 | 14.256123 | 82 | 82 6. SSF Pastoralist Women—Regression ----------------------------------------------------------------------------------------- -> male = Female Source | SS df MS Number of obs = 82 -------------+---------------------------------- F(4, 77) = 11.77 Model | 6246.2485 4 1561.56212 Prob > F = 0.0000 Residual | 10215.9525 77 132.674707 R-squared = 0.3794 -------------+---------------------------------- Adj R-squared = 0.3472 Total | 16462.201 81 203.237049 Root MSE = 11.518 ------------------------------------------------------------------------------ skinsum | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | -.1853649 .0842341 -2.20 0.031 -.3530964 -.0176333 | loc | Lokichoggio | 14.58162 7.932248 1.84 0.070 -1.213502 30.37675 Lorengo | -.2386512 6.86703 -0.03 0.972 -13.91265 13.43535 Lorugum | 24.76832 8.416133 2.94 0.004 8.009656 41.52698 | _cons | 26.29913 8.202305 3.21 0.002 9.966256 42.63201 ------------------------------------------------------------------------------ 104 Appendixes 7. SSF Pastoralist Men—Descriptive Statistics ----------------------------------------------------------------------------------------- -> male = Male Means, Standard Deviations and Frequencies of Sum of Skinfold Thickness (mm) | refliveli | hood2 Location | 0 | Total -----------+-----------+---------- Kakuma | 15.733334 | 15.733334 | 5.5590771 | 5.5590771 | 3 | 3 -----------+-----------+---------- Lokichogg | 14.565 | 14.565 | 2.3193182 | 2.3193182 | 20 | 20 -----------+-----------+---------- Lorengo | 14.084 | 14.084 | 3.057381 | 3.057381 | 75 | 75 -----------+-----------+---------- Lorugum | 14.963636 | 14.963636 | 1.904349 | 1.904349 | 11 | 11 -----------+-----------+---------- Total | 14.306422 | 14.306422 | 2.8994021 | 2.8994021 | 109 | 109 8. SSF Pastoralist Men—Regression ----------------------------------------------------------------------------------------- -> male = Male Source | SS df MS Number of obs = 109 -------------+---------------------------------- F(4, 104) = 2.14 Model | 68.9409785 4 17.2352446 Prob > F = 0.0815 Residual | 838.96454 104 8.06696673 R-squared = 0.0759 -------------+---------------------------------- Adj R-squared = 0.0404 Total | 907.905518 108 8.40653258 Root MSE = 2.8402 ------------------------------------------------------------------------------ skinsum | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | -.0398228 .0155314 -2.56 0.012 -.0706221 -.0090235 | loc | Lokichoggio | -1.082715 1.758821 -0.62 0.540 -4.570523 2.405094 Lorengo | -1.074823 1.687233 -0.64 0.526 -4.420669 2.271023 Lorugum | -.3280262 1.857962 -0.18 0.860 -4.012434 3.356382 | _cons | 17.08731 1.722743 9.92 0.000 13.67104 20.50357 ------------------------------------------------------------------------------ . bysort male: tab loc reflivelihood2 if reflivelihood2==0, sum(bmi)   B Regression of Body Fat Percentage for Women and Men 1. BFP Women reg bfp c.age##i.loc height if male==0, vce(robust) Linear regression Number of obs = 293 F(8, 284) = 9.79 Prob > F = 0.0000 R-squared = 0.2226 Root MSE = 4.873 ------------------------------------------------------------------------------ | Robust bfp | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | -.141199 .0414596 -3.41 0.001 -.222806 -.059592 | loc | Lokichoggio | -8.030881 2.631899 -3.05 0.002 -13.21139 -2.850378 Lorengo | -10.78557 2.428888 -4.44 0.000 -15.56648 -6.004662 Lorugum | 1.340067 3.022918 0.44 0.658 -4.610101 7.290235 | loc#c.age | Lokichoggio | .124541 .0519265 2.40 0.017 .0223314 .2267506 Lorengo | .1489162 .046082 3.23 0.001 .0582106 .2396217 Lorugum | -.0193399 .0572291 -0.34 0.736 -.1319868 .0933071 | height | 2.876207 4.238803 0.68 0.498 -5.46725 11.21966 _cons | 18.41739 7.184531 2.56 0.011 4.275698 32.55907 ------------------------------------------------------------------------------ . margins, at(age=(25 45 65) loc=(1 2 3 4)) Predictive margins Number of obs = 293 Model VCE : Robust Expression : Linear prediction, predict() 1._at : age = 25 loc = 1 2._at : age = 25 loc = 2 3._at : age = 25 loc = 3 4._at : age = 25 loc = 4 5._at : age = 45 loc = 1 6._at : age = 45 loc = 2 7._at : age = 45 loc = 3 105 8._at : age = 45 loc = 4 9._at : age = 65 | Lorugum 1.340067 3.022918 0.44 0.658 -4.610101 7.290235 | loc#c.age | Lokichoggio | .124541 .0519265 2.40 0.017 .0223314 .2267506 Lorengo | .1489162 .046082 3.23 0.001 .0582106 .2396217 Lorugum | -.0193399 .0572291 -0.34 0.736 -.1319868 .0933071 | 106 Appendixes height | 2.876207 4.238803 0.68 0.498 -5.46725 11.21966 _cons | 18.41739 7.184531 2.56 0.011 4.275698 32.55907 ------------------------------------------------------------------------------ . margins, at(age=(25 45 65) loc=(1 2 3 4)) Predictive margins Number of obs = 293 Model VCE : Robust Expression : Linear prediction, predict() 1._at : age = 25 loc = 1 2._at : age = 25 loc = 2 3._at : age = 25 loc = 3 4._at : age = 25 loc = 4 5._at : age = 45 loc = 1 6._at : age = 45 loc = 2 7._at : age = 45 loc = 3 8._at : age = 45 loc = 4 9._at : age = 65 loc = 1 10._at : age = 65 loc = 2 11._at : age = 65 loc = 3 12._at : age = 65 loc = 4 ------------------------------------------------------------------------------ | Delta-method | Margin Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- _at | 1 | 19.58448 1.231203 15.91 0.000 17.16104 22.00792 2 | 14.66712 .749636 19.57 0.000 13.19158 16.14267 3 | 12.52182 .5543328 22.59 0.000 11.43069 13.61294 4 | 20.44105 1.155058 17.70 0.000 18.16749 22.71461 5 | 16.7605 .6359952 26.35 0.000 15.50864 18.01236 6 | 14.33396 .4664967 30.73 0.000 13.41573 15.25219 7 | 12.67616 .3462699 36.61 0.000 11.99458 13.35774 8 | 17.23027 .6349 27.14 0.000 15.98057 18.47998 9 | 13.93652 .8174564 17.05 0.000 12.32748 15.54556 10 | 14.0008 .8022787 17.45 0.000 12.42164 15.57997 11 | 12.8305 .5040656 25.45 0.000 11.83832 13.82268 12 | 14.0195 .8539922 16.42 0.000 12.33854 15.70045 ------------------------------------------------------------------------------   B. Regression of Body Fat Percentage for Women and Men 107 2. BFP Men reg bfp c.age##i.loc height if male==1, vce(robust) Linear regression Number of obs = 284 F(8, 275) = 143.57 Prob > F = 0.0000 R-squared = 0.5571 Root MSE = 2.1066 ------------------------------------------------------------------------------ | Robust bfp | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | .1346334 .020259 6.65 0.000 .0947509 .1745159 | loc | Lokichoggio | -1.516675 1.090777 -1.39 0.166 -3.66401 .6306593 Lorengo | -1.243835 1.136754 -1.09 0.275 -3.481681 .9940105 Lorugum | -.3366657 1.221392 -0.28 0.783 -2.741132 2.067801 | loc#c.age | Lokichoggio | .0124113 .0208331 0.60 0.552 -.0286013 .0534239 Lorengo | .0069982 .0215155 0.33 0.745 -.0353578 .0493543 Lorugum | .0015151 .0231052 0.07 0.948 -.0439705 .0470006 | height | .5281013 1.404562 0.38 0.707 -2.236959 3.293161 _cons | 1.243637 2.818971 0.44 0.659 -4.305868 6.793141 ------------------------------------------------------------------------------ . margins, at(age=(25 45 65) loc=(1 2 3 4)) Predictive margins Number of obs = 284 Model VCE : Robust Expression : Linear prediction, predict() 1._at : age = 25 loc = 1 2._at : age = 25 loc = 2 3._at : age = 25 loc = 3 4._at : age = 25 loc = 4 5._at : age = 45 loc = 1 6._at : age = 45 loc = 2 7._at : age = 45 loc = 3 8._at : age = 45 loc = 4 9._at : age = 65 loc = 1 10._at : age = 65 loc = 2 11._at : age = 65 loc = 3 12._at : age = 65 loc = 4 108 Appendixes ------------------------------------------------------------------------------ | Delta-method | Margin Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- _at | 1 | 5.528265 .609036 9.08 0.000 4.329299 6.72723 2 | 4.321872 .1803343 23.97 0.000 3.966861 4.676883 3 | 4.459385 .2451847 18.19 0.000 3.976708 4.942063 4 | 5.229476 .4006534 13.05 0.000 4.440738 6.018213 5 | 8.220932 .3752184 21.91 0.000 7.482266 8.959597 6 | 7.262765 .1282471 56.63 0.000 7.010294 7.515236 7 | 7.292017 .1356091 53.77 0.000 7.025053 7.558981 8 | 7.952444 .3098849 25.66 0.000 7.342396 8.562493 9 | 10.9136 .4888717 22.32 0.000 9.951192 11.87601 10 | 10.20366 .1712874 59.57 0.000 9.866457 10.54086 11 | 10.12465 .1577277 64.19 0.000 9.814142 10.43516 12 | 10.67541 .3823249 27.92 0.000 9.922758 11.42807 ------------------------------------------------------------------------------         C Worry Codes for Principal Component Analysis 0 No worries 48 Death (children) 10 Drought 49 Disability 11 Thirst 50 Unemployment/underemployment 12 Thirst (children/family) 51 Poverty/hard life 13 Thirst (livestock) 52 Lack of school fees 20 Land for herding 53 Low prices for trade items 21 Lack of livestock 54 Lack of development/services 22 Environmental degradation/ 55 Work too hard for little money depletion 56 Orphan 23 No one to take care of livestock 57 Widow/widower 30 Hunger 58 Lack of shelter/housing 31 Hunger (self) 59 Inadequate education/ 32 Hunger (children/family) literacy issues 33 Hunger (livestock) 60 Raiding 40 Illness 61 Personal safety during raids 41 Illness (self) 62 Child/family safety during raids 42 Illness (children/family) 63 Killings (general) 43 Illness (livestock) 70 Refugee issues 44 Lack of medical facilities 80 Transportation 45 Age 81 Distance/too far to travel for 46 Infertility/fertility problems trade or work 47 Death 90 Dog bites 109 D Regression of Sum of Skinfolds Tests Controlling for Age, Height, and Worries (Total, Economic, and Health) 1. Women reg skinsum c.age##i.loc kids height worriestot i.worriesecon i.worrieshealth if m > ale==0, vce(robust) Linear regression Number of obs = 292 F(12, 279) = 7.19 Prob > F = 0.0000 R-squared = 0.2703 Root MSE = 15.239 --------------------------------------------------------------------------------- | Robust skinsum | Coef. Std. Err. t P>|t| [95% Conf. Interval] ----------------+---------------------------------------------------------------- age | -.5968832 .1274205 -4.68 0.000 -.8477108 -.3460555 | loc | Lokichoggio | -19.47527 8.773328 -2.22 0.027 -36.74559 -2.204947 Lorengo | -30.65133 7.61817 -4.02 0.000 -45.64772 -15.65493 Lorugum | 1.473715 10.06963 0.15 0.884 -18.34838 21.29581 | loc#c.age | Lokichoggio | .3051727 .1732323 1.76 0.079 -.0358356 .646181 Lorengo | .473486 .1440128 3.29 0.001 .1899963 .7569758 Lorugum | .0296817 .1888094 0.16 0.875 -.3419903 .4013536 | kids | .8604609 .429209 2.00 0.046 .0155616 1.70536 height | 5.833894 13.06658 0.45 0.656 -19.88771 31.55549 worriestot | .2128098 .9040241 0.24 0.814 -1.566765 1.992384 1.worriesecon | 5.627722 3.124759 1.80 0.073 -.5233762 11.77882 1.worrieshealth | -2.252239 1.944744 -1.16 0.248 -6.080474 1.575996 _cons | 40.25763 22.82834 1.76 0.079 -4.680018 85.19528 --------------------------------------------------------------------------------- . margins, at(age=(25 45 65) loc=(1 2 3 4)) Predictive margins Number of obs = 292 Model VCE : Robust Expression : Linear prediction, predict() 1._at : age = 25 loc = 1 2._at : age = 25 loc = 2 3._at : age = 25 loc = 3 4._at : age = 25 loc = 4 5._at : age = 45 loc = 1 111 6._at : age = 45 loc = 2 | Lokichoggio .3051727 .1732323 1.76 0.079 -.0358356 .646181 | Lorengo .473486 .1440128 3.29 0.001 .1899963 .7569758 | Lorugum .0296817 .1888094 0.16 0.875 -.3419903 .4013536 | kids | .8604609 .429209 2.00 0.046 .0155616 1.70536 height | 5.833894 13.06658 0.45 0.656 -19.88771 31.55549 worriestot | .2128098 .9040241 0.24 0.814 -1.566765 1.992384 112 Appendixes 1.worriesecon | 5.627722 3.124759 1.80 0.073 -.5233762 11.77882 1.worrieshealth | -2.252239 1.944744 -1.16 0.248 -6.080474 1.575996 _cons | 40.25763 22.82834 1.76 0.079 -4.680018 85.19528 --------------------------------------------------------------------------------- . margins, at(age=(25 45 65) loc=(1 2 3 4)) Predictive margins Number of obs = 292 Model VCE : Robust Expression : Linear prediction, predict() 1._at : age = 25 loc = 1 2._at : age = 25 loc = 2 3._at : age = 25 loc = 3 4._at : age = 25 loc = 4 5._at : age = 45 loc = 1 6._at : age = 45 loc = 2 7._at : age = 45 loc = 3 8._at : age = 45 loc = 4 9._at : age = 65 loc = 1 10._at : age = 65 loc = 2 11._at : age = 65 loc = 3 12._at : age = 65 loc = 4 ------------------------------------------------------------------------------ | Delta-method | Margin Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- _at | 1 | 40.37958 3.900034 10.35 0.000 32.70235 48.05681 2 | 28.53363 2.758078 10.35 0.000 23.10434 33.96291 3 | 21.56541 1.891822 11.40 0.000 17.84135 25.28946 4 | 42.59534 4.038618 10.55 0.000 34.64531 50.54537 5 | 28.44192 2.030559 14.01 0.000 24.44476 32.43908 6 | 22.69942 1.493568 15.20 0.000 19.75933 25.63951 7 | 19.09746 1.169481 16.33 0.000 16.79534 21.39959 8 | 31.25131 2.047205 15.27 0.000 27.22138 35.28124 9 | 16.50426 2.45456 6.72 0.000 11.67245 21.33606 10 | 16.86521 2.97448 5.67 0.000 11.00994 22.72049 11 | 16.62952 1.834828 9.06 0.000 13.01766 20.24139 12 | 19.90728 2.608767 7.63 0.000 14.77191 25.04265 ------------------------------------------------------------------------------       Regression of Sum of Skinfolds Tests 113 2. Men . reg skinsum c.age##i.loc kids height worriestot i.worriesecon i.worrieshealth if m > ale==1, vce(robust) Linear regression Number of obs = 284 F(12, 271) = 2.34 Prob > F = 0.0072 R-squared = 0.0598 Root MSE = 6.3631 --------------------------------------------------------------------------------- | Robust skinsum | Coef. Std. Err. t P>|t| [95% Conf. Interval] ----------------+---------------------------------------------------------------- age | -.0732525 .0624361 -1.17 0.242 -.196174 .0496689 | loc | Lokichoggio | -4.66749 3.099945 -1.51 0.133 -10.77053 1.435546 Lorengo | -4.46278 3.57193 -1.25 0.213 -11.49504 2.56948 Lorugum | -1.621162 3.51388 -0.46 0.645 -8.539136 5.296812 | loc#c.age | Lokichoggio | .0329852 .0622295 0.53 0.597 -.0895296 .1555 Lorengo | .0304762 .0647879 0.47 0.638 -.0970753 .1580278 Lorugum | .0107308 .0686463 0.16 0.876 -.1244171 .1458787 | kids | .0025277 .0659987 0.04 0.969 -.1274076 .132463 height | 1.375086 4.425762 0.31 0.756 -7.338162 10.08833 worriestot | -.0898591 .3772792 -0.24 0.812 -.83263 .6529117 1.worriesecon | -.6929964 1.028033 -0.67 0.501 -2.716942 1.330949 1.worrieshealth | -.4419083 .9529821 -0.46 0.643 -2.318098 1.434281 _cons | 18.87616 8.931464 2.11 0.035 1.29228 36.46003 --------------------------------------------------------------------------------- . margins, at(age=(25 45 65) loc=(1 2 3 4)) Predictive margins Number of obs = 284 Model VCE : Robust Expression : Linear prediction, predict() 1._at : age = 25 loc = 1 2._at : age = 25 loc = 2 3._at : age = 25 loc = 3 4._at : age = 25 loc = 4 5._at : age = 45 loc = 1 6._at : age = 45 loc = 2 7._at : age = 45 loc = 3 8._at : age = 45 loc = 4 9._at : age = 65 loc = 1 10._at : age = 65 loc = 2 11._at : age = 65 loc = 3 12._at : age = 65 loc = 4 114 Appendixes ------------------------------------------------------------------------------ | Delta-method | Margin Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- _at | 1 | 18.75507 1.989338 9.43 0.000 14.83855 22.6716 2 | 14.91221 .6681026 22.32 0.000 13.59688 16.22755 3 | 15.0542 .8116329 18.55 0.000 13.45629 16.65211 4 | 17.40218 1.178911 14.76 0.000 15.08119 19.72317 5 | 17.29002 1.259388 13.73 0.000 14.81059 19.76945 6 | 14.10687 .4041837 34.90 0.000 13.31113 14.90261 7 | 14.19867 .5007608 28.35 0.000 13.2128 15.18455 8 | 16.15175 .8814732 18.32 0.000 14.41634 17.88715 9 | 15.82497 1.527502 10.36 0.000 12.81769 18.83225 10 | 13.30152 .8699758 15.29 0.000 11.58875 15.01429 11 | 13.34315 .5260634 25.36 0.000 12.30746 14.37884 12 | 14.90131 1.11029 13.42 0.000 12.71542 17.0872 ------------------------------------------------------------------------------           Background Paper: E The Future of Kakuma’s Youth— Opportunities and Challenges Context, Rationale, and Methodology Turkana West is characterized by a variety of internal and external pressures, including insecurity, climate change, environmental degradation, and prior marginalization by the state. These stressors have intensified in recent years, challenging the resilience of people, land, and resources in such a way that have turned previously effective coping mechanisms into more permanent adaptations (Huisman 2011). These varying means of survival have had significant implications on how people see themselves, the roles they play in society, and how they relate to others within and outside their communities. Elders, the custodians of the traditional system and culture, have been unable to com- prehensively respond to the evolving pressures. They and the customary laws they rep- resent have conceded much of their authority to state institutions, such as the police and formal courts (Carlson et al. 2012), severely undermining their power and continu- ing to weaken their influence. The distinct roles of men as pastoralists and women as agriculturalists and foragers, divided into public and domestic realms, follow dichotomies that are almost universally observed. In the past, women’s activities were exalted in myth and function. They were acknowledged for their provisioning of food through their expertise in gathering and processing wild plants and micro- and macrofauna and in growing crops such as sor- ghum. However, from the 1900s to the 1980s, changes introduced during the British and Kenyan administrations resulted in significant systemic changes to the social organiza- tion. Women’s labor shifted from playing a central role in the subsistence economy and in myths to being relegated to a secondary role within the emergent awi-adakar system. Ironically, while women continued to maintain their agricultural and foraging practices and to teach younger generations the skills needed for food extraction, processing, and storage as well as how to take care of young animals, their role and status within the larger society diminished. Pastoralism expanded in its central significance beyond sta- tus and prestige, becoming the primary subsistence practice among the Turkana. 115 116 Appendixes The centrality of cattle to the status and livelihood of the Turkana people, their custom- ary institutions, and maintaining social cohesion poses serious challenges. A shortage of livestock undermines the give-and-take social compact that informs traditional cultures (Carlson et al. 2012). This has implications for youth when, for example, they want to marry. Most young men wait a long time before getting married because they are unable to provide bridewealth. In addition, a scarcity of cattle means the loss of a livelihood and an absence of economic security to provide for themselves and their families. Many youth, while acknowledging the importance of pastoralism as a way of life, admit that it is not viable. Education, which has been reinforced by the development narrative as the primary solution to this declining pastoral livelihood, is seen as the only alternative. However, according to the ethnographic findings, one unintended consequence of suc- cessful education outreach efforts in a context that involves structural issues regarding access was the creation of a “lost generation”—a large population of men across Kenya, especially Turkana County, who received a substandard education in a rural or low- income urban area. The drive was high to keep young children in school, but this transi- tion came at the expense of their learning traditional subsistence skills, such as farming and pastoralism. When their education failed to translate into jobs and livelihoods in urban or professional environments, these young men were left without the skills to generate a livelihood in their traditional environments either. Because of this, young men in Turkana compliment traditional pastoralist activities with other socioeconomic activities, such as casual labor, petty trade, road construction, and riding boda-bodas (commuter motorcycles), as examples. Unemployment pushes many male youth into idleness, smoking, drinking, stealing, or engaging in armed raids as a way to deal with rising poverty and hunger. This situation presents a challenge to customary institutions and to the social stability that these institutions are meant to protect (Carlson et al. 2012). Particularly among women and youth, coping strategies to migrate and pursue new livelihoods have resulted in an influx of new ideas and practices that challenge formerly established traditions. This may signal a more permanent shift away from the tradi- tional—something that has profound implications on the viability of the traditional way of life over the long term. The objective of this background paper is to assess the extent to which the peculiar situation of the Kakuma Refugee Camp and Turkana West, as well as the interaction between the two, offer opportunities or pose challenges to realizing young people’s aspi- rations of practicing or leaving traditional livelihoods. The information cited in this paper is gleaned from secondary sources consisting of academic literature, including journal articles and policy documents, complimented by E. Background Paper: 117 The Future of Kakuma’s Youth—Opportunities and Challenges ethnographic data collected for the social impact analysis and from other sources, such as Kenya’s 2009 census. This myriad of information sources provides a broad cross- section of perspectives, observations, and analyses. While most literature focuses on the Turkana, a number of key documents discuss how similar dynamics affect other pas- toral communities, such as the Karamoja and the Maasai. The insights remain relevant because these populations share similar lifestyles, histories, cultures, environments, and methods of production. What is Youthhood? When discussing youth in Turkana West, it is vital to first begin by defining what “youthhood” means to the Turkana, mainly because the concept has a cultural under- pinning that differs from other cultural practices and from the modern definitions of “youth” as stated in official documents, such as the Kenyan constitution, which defines it as all individuals of ages 18–35. The United Nations (UN) defines youth as persons of ages 15–24. Age-group systems are vital among the Nilo-Hamitic peoples. “Turkana boys are initi- ated into full formal adulthood at an average age of 18. The age limits however vary in practice between 14 and 20” (Gulliver n.d. 900). Unlike other Kenyan cultures, the initiation ceremony does not include circumcision. Instead, this right of passage is sig- nified by a boy spearing a castrated male animal—ox, camel, goat, or sheep—given to him by his father at a communal ceremony. Afterward, the carcass is opened, and the boys’ head and body is smeared with the undigested contents of the stomach by the most senior elders of the area. The most senior men in the age-group play a critical role in receiving the initiates into the system and passing on to them the attributes of strong manhood and the idealized qualities of age-group membership (Gulliver n.d.). Once initiated, a young man is able to wear a man’s mudded headdress and has the right to hold a proper spear instead of the wooden sticks used by boys. A youth is assigned to an age-group and thereafter participates in specific age-group activities, such as feasting, dancing, and war-making. He is consistently guided by his affiliation to that age group, which fosters general feelings of association and attitudes of assistance and mutual support. The young man is allowed to marry, but in practice may not be able to for several years depending on his position in the nuclear family and whether or not the size of his family’s herds allow him to pay a bride-price. The average age for first marriage4 among Turkana men is currently 31.8 years (Ohta 2007). 4. Marriage in African cultures is the ultimate transition from youth to adulthood. 118 Appendixes Table E.1. Drought Magnitude in Lodwar, Turkana For women, marriage is the first and primary stage (1950–2008) marking the important transition from childhood Livestock to adulthood. Turkana girls are usually married Drought Mortality between ages 15–20,5 although some studies (for Magnitude Rate example, Ohta 2007) report an average age of 22.7. Year (1950–2008) Name of Drought (percent) 1950 -0.43 References to “youth” in Turkana West in this 1952–57 -2.84 Lotiira 61 paper loosely refer to 14- to 32-year-olds regard- 1960 -0.65 Namotor 55 less of marital status or the definition in the Kenyan 1964 -0.18 constitution. 1971 -1.09 Kimududu 54 1976 -0.27 Kibekbek 54 Changes Since the Setting up of 1980–84 -6.25 Kiyoto atang’aa 65 Kakuma Refugee Camp in 1992 1986–87 -0.28 Lopiar 1990–95 -6.0 Lokwakoyo/Akalkal 53 Kakuma is located in Turkana County, Kenya, 1999–2001 -3.31 Logara/Epompo 63 which lies between longitude 34o0’ and 36o40’E and 2005 -0.06 between latitude 10o30’ and 5o30’N. About 96 per- cent of the county falls under the arid or very arid 2008 -0.59 ecoclimatic zones, with primarily dry thornbush a. Drought magnitude is the cumulative sum of the consecutive negative standard precipitation index values for a number of years. Standard precipitation land and dwarf-shrub grasslands. Rainfall patterns index values of -2.0 and less are extreme. Values of -1.0 to -1.5 indicate moderate drought. and distribution are unreliable and erratic, with an Sources: IDRC, Adaptation, and DFID n.d. and ILRI and VSF-B n.d. annual average of 430 millimeters (mm), for the most part peaking in April and November (Ohta 2007). The daily temperatures range from 24oC to 38oC (GOK 1997) and, therefore, potential evapotranspiration rates in excess of 2,500 mm/year are typical (Ekaya 1998). The main economic activity in the region is nomadic pastoralism. Increasing Environmental Vulnerability Climate data confirms that Turkana West experienced a mean rainfall decrease of 13 mm between 1973 and 2008 (Ogindo et al. 2009). The frequency and severity of droughts have increased in recent decades, with episodes of moderate to severe drought occur- ring more frequently (every 2–5 years as shown in table E.1) since the 1980s. Lodwar experienced its most severe drought between 1980 and 1984. The 1990–95 Lokwakoyo/ Akalkal drought was equally severe (see table E.1), causing the loss of more that half of the livestock (ILRI and VSF-B n.d.). The setting up in 1992 of Kakuma refugee camp in Turkana West coincided with the Lokwakoyo/Akalkal drought, and studies show that the camp has had a negative impact 5. See “Turkana—Society” at website of Jens Finke at http://www.bluegecko.org/kenya/tribes/turkana/society.htm. E. Background Paper: 119 The Future of Kakuma’s Youth—Opportunities and Challenges on the ecological integrity of the area (Okoti et Table E.2. Tree-Crown and Shrub Cover by Distance al. 2004). In a 2004 study conducted in the area from the Kakuma Refugee Camp (Okoti et al. 2004), there was a significant dif- Distance from camp (km) 1 2 3 ference in tree-crown cover along the distance Mean tree crown cover (percent) 6.2 18.8 28.7 gradient. As shown in table E.2, there was low Mean shrub cover (percent) 0.9 6.9 9.8 tree-crown cover near the settlement camp, Source: Okoti et al. 2004. which increased with distance from it. Trees near the camp that had been an important source of forage for livestock were cut down for burning charcoal and construction. The chop- ping down of important trees such as Acacia tortilis denies pastoralists the tree pods that serve as feed during the dry season, and the resultant bare soils are vulnerable to wind and water erosion. Table E.2 also shows an increase in shrub cover as distance from the camp increases. The need for fencing and building materials exerted the most pressure on these shrubs. The availability of forage lands in the area was also affected by changes in rainfall and an increase in the population. The keeping of livestock around Kakuma throughout the year meant that a range of resources was overutilized, exposing this fragile ecosystem to erosion. Among the people interviewed, 65 percent kept their animals near the camp throughout the year instead of moving them to other areas in search of forage, as is the nomadic practice. The number of animals surpassed the availability of fodder, causing further deterioration of the land. Loss of Livestock Livestock plays a central role in Turkana, not only as a symbol of status and prestige, but also as the lifeline and identity of a Turkana man. One refugee man from the Oromo community explained: “When you see a Turkana man like that, sitting there, not doing anything, on his stool, holding his stick, you think, why doesn’t he work? But you know from looking at him that he has lost his herd. … That man, there, he is broken. He is broken, because when he lost his herd, his animals, he lost himself. He may start drinking, start abusing drugs, get into fights, but the biggest problem is that he lost his herd, and he lost his pride, his status” (KII 2010). Second, the value of livestock goes beyond the production of meat to the supply of milk and blood for consumption. The exchange or sacrifice of animals underpins every cere- mony and ritual and, as such, the customary judicial systems. Under customary author- ity, animals are meant to be paid as compensation for wrongs committed, sacrificed as 120 Appendixes Table E.3. Livestock Population in Turkana District (1993–2009) Percentage Total Number of Livestock of Livestock in Population in  Year Cattle Camels Sheep Goats County 1995 1993 153,350 63,153         1994 165,000 94,000         1995 198,000 112,800 862,000 1,704,000 2,876,800 1.00 1996 201,960 114,192 894,333 1,788,667 2,999,152 1.04 1997 200,000 115,230 916,667 1,833,333 3,065,230 1.07 1998 200,000 115,000 916,667 1,833,333 3,065,000 1.07 1999 234,420 144,960 1,084,050 2,168,100 3,631,530 1.26 2000 176,000 138,000 813,000 1,626,000 2,753,000 0.96 2001 176,000 138,000 813,000 1,626,000 2,753,000 0.96 2002 193,600 140,760 975,600 1,951,200 3,261,160 1.13 2003 193,600 140,760 975,600 1,951,200 3,261,160 1.13 2004 197,700 172,400 1,054,400 2,021,000 3,445,500 1.20 2005 197,900 172,400 1,054,400 2,021,000 3,445,700 1.20 2009 1,534,612 832,462 3,519,148 5,994,881 11,881,103 4.13 Source: GOK 2002 and KNBS and SID 2013. a means of making amends, and exchanged in bonds of social reciprocity that secure relationships, heal old wounds, and establish a guarantee of assistance during difficult times in the future (Carlson et al. 2012). The loss of animals over time has rendered the pastoral livelihood of the Turkana and many aspects of their customary law ineffective. While livestock numbers in pastoralist areas are difficult to measure, table E.3 illustrates the yearly change in Turkana County. The data challenge the common assertions that pastoral livestock populations are either exploding or imploding. There is evidence that livestock per capita increased in the late 1980s but crashed in the following years and that the average livestock per capita was far below the recommended values for subsistence requirements (4.0–7.1 livestock per capita) (Ebei et al. 2007). Livestock numbers fluctuate considerably due to uncertain environments. Increases and decreases of around 50 percent or more over a year or so are not uncommon—pastoral livestock are highly prolific in the immediate aftermath of climatic shocks (Davis 2007). E. Background Paper: 121 The Future of Kakuma’s Youth—Opportunities and Challenges Using average mortality rates of 50 percent for cattle, 30 percent for sheep, 24 percent for goats, and 17 percent for camels, the estimated recovery periods are 10, 2, 1.5, and 4 years, respectively (ILRI and VSF-B n.d.). Such long recovery periods leave pastoralists vulnerable because herd growth can be disrupted by subsequent droughts before the recovery is complete (ILRI and VSF-B n.d.). Severity of droughts and their impact on livestock production also translate into reduced purchasing power for pastoral house- holds. Because of this, poverty levels in Turkana rose from an estimated 61 percent in 1999 (CBS and ILRI 1999) to 88 percent in the 2009 census (KNBS and SID 2013). The ethnographic data collected suggest that the growing cultural investment in pas- toralism as the primary activity among men, combined with declining herd sizes well beneath the asset threshold necessary for resilience or to buffer against shocks, may explain the sense of general despair experienced by many Turkana men who find themselves without a purpose or status in their society. While the exclusion of many men from generating sizeable herds and gaining status could open up other opportuni- ties, the cultural insistence on pastoralism renders alternatives as lacking in honor or respect. One man explains the problem of Turkana men who no longer have status through successful herding: “Without animals, a man cannot be a Turkana man. It is not easy to build a herd from nothing. If you have sisters, you can get animals through dowry. But then you have to get enough animals for your own dowry…. Even the ones who have jobs, they buy animals so that when they retire, when they’re old, they can be pastoralists.” (KII 2011) Increased Migration and Population The setting up of Kakuma camp coincided with the Lokwakoyo/Akalkal drought. Many people within and outside Turkana initially moved to the vicinity of the camp in search of forage and water for their animals in addition to relief food for themselves, which was available at the Kakuma Refugee Camp. In 1989, before the creation of the camp, Kakuma was a small town of slightly more than 2,000 people. Its population grew to more than 9,000 by 1999 (Ohta 2005). By 2009, the population in Kakuma ward was 33,539 people (KNBS and SID 2013) exclud- ing the refugee community. A study conducted in 2004 found that 50 percent of the locals living in Kakuma were born there and another 50 percent were from the sur- rounding regions. The study also found that about 65 percent of the people who settled in Kakuma came during 1990/95 drought (Okoti et al. 2004). 122 Appendixes Embracing Alternative Livelihoods According to oral histories, archaeological evidence, and the ethnographic data of this study, the pre-1900 social organization of the Turkana was closely aligned with the diversity of the subsistence pathways among the people. Activities of men as pastoral- ists and women as agriculturalists and foragers were divided into public and domestic realms following almost universally observed dichotomies. However, women’s activities were exalted in both myth and function as they were acknowledged for their provision- ing of food and for food security through their expertise in gathering and processing wild plants, micro- and macrofauna, and in growing crops such as sorghum. The changes introduced during both the British and the Kenyan administrations, from the 1900s to the 1980s resulted in significant systemic changes to this social organiza- tion wherein women’s labor shifted from playing a central role in both the subsistence economy and mythos to being relegated to a secondary role within the emergent awi- adakar system. Ironically, in this development, while women continued to maintain their agricultural and foraging practices and to teach younger generations skills of food extraction, processing, and storage, and also to take care of the young animals, their role and status within the larger society diminished. Pastoralism expanded in its central significance beyond that of status and prestige, becoming the primary subsistence prac- tice associated with the Turkana. This change was paralleled by forced settlement of the Turkana from 1930 to 1980 into small settlements, which became the boundaries of the domestic realm for women. Interestingly, the ethnographic data of this study shows contemporary Turkana women’s adaptability in diversifying their subsistence pathways and mirrors the actions of their ancestors in their continuing interest in farming and foraging, especially among the rural awis. The data also shows that Turkana women are the primary source for their household’s income and food security. At the onset of the camp, most locals did not have any other source of subsistence except their livestock, which accounted for 90 percent of employment and more than 95 percent of family incomes and livelihoods. Because of the severe drought, families started looking for alternative sources of income. The women in the study reported a diverse range of livelihoods, including agricul- ture; foraging for wild plants such as edapal, elamach, edu, and acacia; making and selling charcoal; collecting and selling firewood; gathering thornbush and poles for E. Background Paper: 123 The Future of Kakuma’s Youth—Opportunities and Challenges fence-making and construction; making fences and walls in the camp; working in the homes of refugees; performing menial cleaning jobs in the development/relief workers’ compounds; and even engaging in commercial sex work. Many women have used the shops in the camp to buy goods to sell at retail prices in the settlements outside the camp where they live and have also built small businesses selling tobacco, relief food, milk, and other consumables to refugees and the Turkana alike. These opportunities have prompted host communities to remain around the camp. From Prominence to Decline: The Role of the Elders The possession of livestock plays multiple social, economic, and religious roles in pas- toral livelihoods, such as providing a regular source of food in the form of milk, meat, and blood for household members; as well as cash income to pay for cereals, education, health care, and other services. In pastoral communities, livestock is also essential as payment of a bride-price; compensation of injured parties during raids; a symbol of prosperity and prestige; and a store of wealth and security against drought, disease, and other calamities. Livestock is therefore a fundamental form of pastoral capital in addi- tion to functioning as a means of production, storage, transport, and transfer of food and wealth (Behnke 2008). The decline of livestock as a result of drought and disease profoundly affected the gov- ernance of elders, resulting in their having much less influence over or relevance to people’s daily lives. This trend has impacted the functionality of customary practices and removed the basic means for ceremony and ritual, as well as the basis for recon- ciliation, compensation, external negotiation, and redress (Carlson et al. 2012). In the absence of compensation in the form of a fine paid by the perpetrator, judicial rulings lost their weight. Amends without a shared feast following an animal sacrifice lacked the weight to mend rifts created by wrongdoings. Offerings of respect made by youth to elders in the form of animals, which meant that elders had influence over the actions of the youth, became less frequent. Therefore, the elders’ ability to shape, mitigate, or manage the acts of young men began to shrink. The transition into other modes of livelihood, such as casual labor and petty trade occa- sioned by a significant loss of pastoral livelihood and the opportunities provided by the refugee camp’s presence also proved extremely challenging to elders, mainly because they did not historically exercise any ritualized authority over these new forms of liveli- hoods (Carlson et al. 2012). 124 Appendixes Declining Social Network Pastoralist communities tend to have very high social capital6 because association is considered to be at the core of all livelihood strategies—even more than livestock. The society is therefore organized in a way that provides structural links (through gener- ation-sets and age-sets) and dynamic links (through kinships and friendships), cre- ating economic interdependencies and safety nets. These links are cemented through the horizontal exchange of livestock. Limited numbers of livestock therefore mean that people are less able to establish and maintain a larger social network of alliances and friendships that allow for the mitigation of vulnerability. The system is further inhibited by increased insecurity and sporadic theft of livestock (Carlson et al. 2012). From Cultural Raiding to Its Commercialization The creation of the refugee camp brought with it a high demand for meat among the ref- ugees, necessitating more frequent supplies of meat. Internal theft of livestock escalated to satisfy the demand. Cattle raiding—previously done for reasons such as articulation of hostility toward enemy communities, restocking of best livestock to replace animals lost during periods of drought or disease, expansion of grazing lands, gaining access to water and pasture resources, and increasing social status (Eaton 2008)—became more frequent to meet the increasing demand for meat. With the change in its purpose from cultural to market reasons, raiding moved from mass invasions comprising more than 100 raiders communally organized by elders (an adakar) to raids by smaller groups of people—mainly male youth under 30 years old (Schilling et al. 2012). A study conducted by TUPADO suggests that improved commu- nication enabled smaller groups of raiders to easily organize discreet raids, and therefore between 2006–09, the number of raiders decreased from 48 to 28 per attack (TUPADO 2011). Raiders targeted large unprotected herds, especially those in the care of young boys. The raids, formerly fought with spears, bows, and arrows, were now fought with AK-47s acquired from neighboring countries experiencing civil wars, increasing the level of violence and fatal outcomes resulting from the raids (Krätli and Swift 2003). The increased integration of the pastoralist economy with the wider market economy therefore incentivized more violent, frequent, and well-organized raids. Raiding was undertaken with the explicit intention of selling livestock for immediate profit instead of restocking one’s own herds, as had previously been the case (Mkutu 2010). An increasing 6. Social capital, as defined by Putnam 1993, refers to features of a social organization, such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit. E. Background Paper: 125 The Future of Kakuma’s Youth—Opportunities and Challenges urban population and improved infrastructural connectivity between arid areas such as Turkana and the rest of the country further intensified the raids, evidenced by the fact that Kenya’s drylands have over 50 percent of the country’s livestock population and produce over 67 percent of the red meat consumed in the country (Davies 2007). Drivers of Change and Their Resultant Effects Changes witnessed in Turkana, and more specifically in Turkana West, are attributed to the following key drivers. Education Education has been reinforced by the development narrative as the primary solution to a decline in pastoral livelihoods. There is a saying that “education does not know drought” (Carlson et al. 2012). The Turkana also recognize that education empowers communities, links people to the broader world, and develops the human capital neces- sary for sustainable development. Education is therefore pursued as an escape route from pastoralism. It is perceived as a safety net or a way of accessing resources outside the pastoral circuit. Education is also viewed as a way to decrease vulnerability and supplement the family’s eroding source of livelihood due to drought, cattle rustling, or other reasons (World Bank 2001). The rise in school enrollment over the last few years (see figure E.1) is likely a testament to this fact. However, while education is emerging as an adaptive response among Turkana West youth to access new livelihoods, most cannot afford to pursue a formal education. Figure E.1. Total Primary School Enrollment (Class 1–8) by Gender in Kakuma, 2000–08 60 50 Boys Girls 40 30 20 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 Source: Migosi et al. 2012. 126 Appendixes Table E.4. Education by County, Constituency, and Ward in Free primary education has improved Turkana West (percentage) enrollment rates at the primary level Secondary Total (figure 1). However, education levels County/Constituency/Ward None Primary or Higher Population among people in Turkana West are Kenya 25.2 52.0 22.8 34,024,396 still very low. Only 8.5 percent of the Rural 29.5 54.7 15.9 23,314,262 population had primary education Urban 15.8 46.2 38.0 10,710,134 compared to a national average of 52 Turkana County 82.1 14.5 3.3 749,235 percent (table E.4). Kakuma Ward Turkana West Constituency 89.2 8.5 2.3 182,679 has the highest proportion of indi- Kakuma Ward 72.1 22.5 5.4 30,373 viduals with a primary education Lopur Ward 91.4 5.6 3.0 13,521 at 22.5 percent, mainly because of Letea Ward 98.7 1.2 0.1 41,288 the presence of the camp. However, Songot Ward 96.3 3.3 0.3 16,682 the proportion of individuals with a Kalobeyei Ward 94.6 4.9 0.5 17,226 secondary education drops fourfold in Kakuma Ward, which could be Lokichoggio Ward 73.0 19.3 7.7 24,280 explained by a number of factors, Nanaam Ward 96.4 3.2 0.4 39,309 Source: KNBS and SID 2013. particularly affordability given the levels of poverty in Turkana, and a severe shortage of facilities. Accord- ing to CRA (2011), Turkana County has 202 primary schools and only 19 secondary schools. While the national literacy rate for adults in Kenya is 87.4 percent, only 18.1 percent of Turkana County’s population can read and write. The ethnographic data collected for this study shows that most respondents cited struc- tural barriers that prevent many people from receiving the benefits of education. The responses reveal that the lack of adequate educational infrastructure in Turkana is a complex phenomenon, which includes: n An overall lack of schools and educational institutions; n A disproportionate number of primary schools compared with secondary and ter- tiary educational institutions; n Low numbers of qualified teachers at all levels; and n Inadequate funding for public schools, resulting in lower salaries for teachers and a lack of educational materials. Public schools do not charge school fees, but students still incur some costs. A failure to pay can result in a student being expelled or forbidden to attend schools until the debt is E. Background Paper: 127 The Future of Kakuma’s Youth—Opportunities and Challenges paid. Private schools charge for materials, fees, boarding, and lodging. These structural problems are summed up by one woman who said: “The children who are brought into town by parents, because leaders say bring them, even though the government is providing free schools, they have to pay exam fees, uniforms, books, food. But the teaching is a problem. Many of my friends who are teachers say that they get no support for teaching. So these young people end up on the street, and do jobs here and there, in town, in camp. They end up going the wrong way many times. So when the pasto- ralists see this, they say, that our herding, that is the good life. But they are stuck in that life. Now many of us are saying, let some of us get schooled and some do herding and pastoralism. […] I think that some education is good but maybe, now that so many people are exposed to business, and even do business among the Turkana, maybe that is what they need—business educa- tion. One man here—I’ll take you to him—he owns lorries, rental houses, hardware stores. He was a pastoralist and now is very rich. But he never went to school.” (KII 2015) The ethnographic data also demonstrate an interdependent relationship between pushing for education and maintaining close contact with rural kin. Almost all of the respondents living in Kakuma town or near the camp have relationships with members of their awi, which includes urban relatives giving commercial goods to their rural kin, especially in times of drought, and the rural herders giving their urban kin goats or live- stock if sudden infusions of cash are needed, especially to pay for school fees, uniforms, books, and travel costs, as examples. In one woman’s words: “… people, the big ones, they keep telling us to send our children to school, I agree. But the schools should be better. Even though they are free, the children have to pay for uniforms, for books, and we sometimes don’t have enough to eat, so getting money for those things, we have to go to our relatives and get goats to sell. So even if we do what the government wants us to do, because we don’t have the money to do education properly, we need the pastoralism. But then they tell us not to do that. So where will we get the money to buy the things our children need to get a good education? This the leaders do not say. And if the pastoralists are neglected and they don’t have goats, because of drought or famine, they have sold them in the markets just for buying food, then what should we do? If the government wants us to go for education and schooling, then they should provide the services that make that possible.” (KII 2015) 128 Appendixes One man describes the process, especially in relationship to education: “… even though they say that herding is not good, that is not how it is here. […] The political leaders, the teachers, the [nongovernmental organization] workers. They say we must go to school and not do herding. But this is our way, when times are good, when we need money for school, for clothes, we go to our brothers and fathers, and mothers out there, who are herding the cattle and they give us goats to sell, to pay for our things. When things are bad, like food shortage and drought, we send food and other things to them. That is how we contribute.” (KII 2015). An educated population heralds a diversification of livelihood strategies, which could ultimately support pastoral livelihoods by decreasing pressure on animal resources, but older people believe that education erodes cultural identity and is sometimes antipasto- ral. These sentiments are confirmed by the data presented in table E.5, where the level of youth involvement in family livestock holdings decreases with increasing levels of education. In Turkana West, only 8.6 percent of youth with a secondary education are engaged in family livestock holdings compared with 16.5 percent of youth with a primary educa- tion and 48.4 percent with no education. This can be explained by the fact that the higher the education level among young people, the more likely they are to look for and acquire paying work. The fact that the average proportions of engagement with livestock in Turkana West are lower than those of Turkana County in general (55.4 percent for those with no education; 27.3 percent for those with a primary education; and 15.3 percent for those with a secondary education or above) can also be attributed to alternative livelihood opportunities available in the camp or as a result of it. The data also supports the notion that young people with an education were more likely to acquire wage-paying employment than their uneducated peers. A considerable number of youth in the constituency with some level of education (15 percent with secondary education and 22 percent with primary education) have no work. The Turkana, according to ethnographic data collected for this study, attribute this to the general discrimination against the Turkana by the rest of the Kenyan society and the less-than-minimal public infrastructure for education, which may leave Tur- kana youth in a position where they cannot gain status through salaried professional jobs within or outside Turkana County. As one man said: Table E.5. Type of Work Done by Level of Education County/ Educational Work Family Family Agricultural/ Intern/ Retired/ Full-time No Number of Constituency Level for Pay Business Livestock Holdings Volunteer Homemaker Student Incapacitated Work Individuals Turkana Total 5.6 16.9 50.1 1.5 6.5 4.0 0.5 14.9 422,657 Turkana None 3.4 17.5 55.4 1.5 6.6 0.1 0.5 15.1 353,488 Turkana Primary 9.8 15.2 27.3 1.7 7.7 23.3 0.6 14.4 44,792 Secondary or Turkana higher 28.8 11.7 15.0 2.3 4.0 24.6 0.2 13.4 24,377 Turkana West Constituency Total 5.8 12.8 44.7 2.4 8.8 2.4 0.6 22.6 108,785 Turkana West Constituency None 3.7 12.8 48.4 2.4 9.2 0.1 0.6 23.0 97,009 Turkana West Primary 12.8 12.9 16.5 2.6 7.3 24.8 1.3 21.8 7,604 Constituency Turkana West Secondary or 41.4 13.6 8.6 2.4 3.5 15.3 0.3 14.9 4,172 Constituency higher Source: KNBS and SID 2013. The Future of Kakuma’s Youth—Opportunities and Challenges E. Background Paper: 129 130 Appendixes “The local schools are very bad, and do not prepare you for jobs in Nairobi. Even if we are qualified, we won’t be given the jobs, because we are Turkana. But those who went to school, they don’t know herding, so they come, and get small jobs, working for this or that.” (KII 2011) A woman said: “If our government can help us to increase our standards of living, to give the same services and help that the refugees get that will be good. I do not blame the refugees for what they get. They suffer hardship. But this is our home and we are neglected by our own people. Even good jobs here, they go to the down-Kenyans, the Kikuyus, and others. They build the shops here, the Somalis and Kikuyus and Luhyas. We deserve more than we have been given.” (KII 2015) One unintended consequence of increased enrollment and the structural issues in accessing education has been the creation of a “lost generation,” referring to men who received a substandard education. A lack of relevant skills combined with the failure of the system to translate education into jobs and livelihoods in urban or professional settings have left these young men without the ability to generate a livelihood, either in their more traditional lives or by establishing careers in the professional world. While this affects both men and women, the broad division between gendered spaces allows women to gain and maintain status through marriage, motherhood, and domestic housework (female domestic spheres), while men find it impossible to maintain or gain status without engaging in wealth-creation activities in the public sphere, either in the traditional realms of farming and pastoralism or in professional careers. These assertions about education and pastoralism concur with studies around the world, which suggest that although formal education may successfully prepare children for new forms of livelihood outside pastoralism, it is antagonistic to pastoral livelihoods virtually everywhere. The Constitution of Kenya and Its impact on Customary Laws and Practices Customary law encompasses the unwritten norms and practices of small-scale com- munities that pre-date colonial times but which has undergone a number of transfor- mations due to colonialism and capitalism. It is localized in nature and as diverse as the communities involved, although there is general consensus on certain fundamental principles: the central role of the family, the supremacy of the group over the individual, and the importance of kinship ties (Kamau n.d.). E. Background Paper: 131 The Future of Kakuma’s Youth—Opportunities and Challenges Customary laws and practices have an intricate link with the rights of young women. This is especially true because traditional African societies are governed on the basis of patriarchal structures under which the individual interests of women are subsumed by those of the group. Practices such as gender-based violence and early or child marriages are common among the Turkana. These practices not only negatively affect the health, social welfare, dignity, and physical and psychological development of young women, but they are also in contravention of the law as espoused in the Kenyan constitution and international human rights instruments. Section 53(1)d of the constitution, for example, provides that “every child should be protected from abuse, neglect, harm- ful cultural practices, all forms of violence, inhuman treatment and punishment, and hazardous or exploitative labour.”7 Section 55d stipulates: “the State shall take measures, including affirmative action programmes, to ensure that the youth—are protected from harmful cultural practices and exploitation.” And Section 45, subsection 2 specifically stipulates: “every adult has the right to marry a person of the opposite sex, based on the free consent of the parties.” The reference to “adult” presupposes that the individual must have attained the age of 18, effectively outlawing marriage among minors, and the reference to “free consent” outlaws forced marriages. The conflict between the law and human rights practices with customary laws and prac- tices necessitate change because the constitutional provisions are supreme. Urbanization Refugee camps are often perceived as transient settlements, reflecting the temporary nature of the refugee phenomenon. Yet some camps, such as Kakuma—which has existed for 23 years—can be compared to a virtual city given the size of its population and demographic density (Kagwanja 2000). With over 160,000 refugees living there in September 2016, the camp has emerged as an urban enclave. The establishment of infrastructure and social amenities by NGOs and the UNHCR has been valuable for economic development and has benefited both the refugee and local population. Kakuma Refugee Camp contains kindergartens, primary and high schools, vocational schools, a hospital, clinics, libraries, community centers, churches, and mosques. People engage in business at restaurants, general stores, butcher shops, and vegetable stands. Theaters show movies and promotion videos for popular singers, such as Michael Jack- son. Several restaurants have satellite broadcast dishes so people can watch soccer games taking place in Italy. There are telephones and remittance services with international 7. The 2010 Kenyan constitution is available at: https://www.kenyaembassy.com/pdfs/the%20constitution %20of%20kenya.pdf. 132 Appendixes capability. Boda bodas busily come and go on the roads of the camp, undoubtedly presenting many aspects of modernity to the residents. Because of the camp’s growth, Turkana West, which was initially merely a host region for refugees, grew into a large town (Kurimoto 2002). Introduction of a Cash Economy Trade, including the sale of wood for fuel, cutting and selling building poles and fenc- ing material, collecting and selling stones for construction, and selling milk and meat, in addition to employment opportunities, introduced the community to a previously nonexistent cash economy. In a survey conducted in 2004 (Okoti et al. 2004: 18), 95 percent of respondents said they received market value for what they sold, and 77.5 percent said they made a profit, although they could not define it in monetary terms. Profits were spent on food, clothing, medicine for their livestock, for themselves, or for their families, and restocking—something that used to be much more difficult to do. The price of livestock jumped as a result of the high demand for meat from the refugee community, further accelerating the spread of a cash economy among the Turkana. Effects of Changes on Youth Perceptions, Perspectives, and Aspirations The transformations in Turkana West that resulted from the presence of the Kakuma camp set in motion changes that affected the economic, social, and cultural setup of the local community (Okoti et al. 2004). It would seem that Turkana West youth have also changed to accommodate the new reality (Kamau n.d.). Some of the changes associated with the host youth are discussed below. Changing Perception of Pastoralism as a Sustainable Way of Life Mainstream society in Kenya represents pastoralism as a fundamentally flawed way of life. The public image of pastoralism painted in the media, in the education system, and in the public administration has created a public and institutional image that grossly misrepresents it (World Bank 2001). “At the core of the public representation of pastoralism is the idea that ‘pas- toralism’ and ‘modern life’ are mutually exclusive, as two successive stages of human development in a unique line that goes from nature to civilization, passing from sedentary life and agriculture. This frame offers no ground on which pastoralism and modern world could meet: one being thought to begin E. Background Paper: 133 The Future of Kakuma’s Youth—Opportunities and Challenges where the other is supposed to end. The discourse about pastoralists and pastoral development is therefore constructed along strings of oppositions: nature vs. civilization; nomadic vs. sedentary; traditional vs. modern; igno- rance vs. education; irrational vs. rational; dirty vs. clean; women’s subjuga- tion vs. gender sensitivity; group tyranny vs. individual freedom; prosperity vs. poverty; etc.” (World Bank 2001). Young people in Turkana perceive their way of life as being inferior and thus shun it for the “modern,” which is thought to be superior. As indicated by the ethnographic data, there have been efforts to supplant the Turkana’s ways of knowing with those of mainstream Kenyan society. During the late 1960s and 1970s, young Turkana boys and girls were forcibly removed to schools in down-Kenya, punished for speaking the Turkana language, and given an education under stark and often abusive conditions in an attempt to acculturate and reeducate the Turkana into mainstream Kenyan society. These actions were paralleled by the continued enforce- ment of the “closed district” policies of the Kenyan government. A young Turkana woman whose father had been taken to down-Kenya explained: “My father was born in a pastoral family. But he was forced to leave the awi after independence, and [the government] wanted to change their life through education so they can be given jobs in ‘wazungu shambas’ [the white settlers’ ranches]. About the forced settlement, my father said he is okay with the forc- ing now, but it was pretty bad then. The officers would come and take boys and some girls, and put them in schools in down-Kenya. And there were good things because he became a doctor, and most other people who took advan- tage at that time, they became politicians and teachers. But many resisted, and were harassed. Most of the people who did not accept, were beaten, many died in the schools. But those who were humble, they did well. [By humble, I mean] the ones who did what the teachers told them, who did not resist the officers, they did well” (KII 2015). Such stories feed the larger narrative among the Turkana that their culture is considered inferior by mainstream Kenyan society and that if they want to participate in the larger society, they need to assimilate, minimizing their “Turkana behaviors” in their daily lives and activities. 134 Appendixes Changing Perspective of Nomadism Because they are of pastoral origin, local youth used to be indifferent to borders and a sense of permanence. However, the harsh unsustainability of pastoralism and the new- found trade, education, and employment opportunities around the camp in Turkana West have compelled youth to settle in the area (Kagwanja 2000). Embracing of New World Views The gradual expansion of external influences, such as the refugee camp and people from other parts of the country, is an important factor in thinking about changes to traditional structures. Increases in migration and population size in Turkana West intensified interactions among the local community with the migrant community from other parts of the coun- try, aid workers, and refugees. Apart from expanding the networks of young people beyond their local area, the cosmopolitan make-up brought an inflow of new ideas and practices. While these did not automatically result in the erosion of customary author- ity (Carlson et al. 2012), they began influencing customary institutions to evolve. The interaction played a major role in challenging the sedentary lifestyle of pastoral youth. As observed by Betti (2010), generational change is a phenomenon pertaining to all cultures, and therefore there is nothing new about youth being different than the elders in thought, belief, and practice. However, because of changing circumstances, younger generations must resocialize in order to adapt to or make sense of the new world around them. While youth may seem to have drastically changed from their ancestors to fit into new social parameters and value systems, they are really just adjusting their strategies of behavior. For example, youth in Kakuma increasingly prefer to dress in jeans and modern clothes, which can be viewed as assimilation into the “global taste” (Kurimoto 2002). Change in Gender Perspectives and Roles Turkana women have traditionally played a negligible role in decision making. How- ever, due to the decrease of animal populations and the corresponding decline in economic roles for men, most households have adopted a mixed strategy. One foot remains in pastoralism, mainly maintained by the men, while the other is planted in alternative livelihoods that are mainly taken up by the women. With the new role of women supporting their households through resource exploitation and petty trade— such as collecting firewood, burning charcoal, getting milk to sell at the camp, mend- ing fences at the camp, and providing domestic services to refugees—the gender biases embedded in customary law are slowly waning. Young women, backed by the formal E. Background Paper: 135 The Future of Kakuma’s Youth—Opportunities and Challenges law, are challenging cultural practices, such as gender-based violence and early or child marriages. This tension has been newly animated by the increased role of women in economically providing for their households (Carlson et al. 2012). However, this economic transition from traditional livelihoods, while welcomed by young women, puts pressure on them to work “double shifts,” and while these new opportunities may generate more household income or a better livelihood, they may also disempower young women if they result in the women bearing additional burdens and experiencing heightened vulnerability (Enns and Bersaglio 2015). According to the ethnographic data collected for this study, changing gender dynamics have heightened gender-based violence. However, while these new jobs do expose women to violence, including sexual assault, assault, and robbery, the women view the violence as a part of their lives, as a normal occurrence, and the risk as necessary. As dangers mounted, refugee women were discouraged from leaving the camp, but the increasing demand for firewood meant that local people had to supply the camp with the necessary goods. However, taking over the market also included taking over the associated risks because the gangs and militias were still around. While these activities make women more vulnerable to physical violence and even health problems resulting from carrying large quantities of food or water, they also provide a more stable means of income for women, ranging from approximately K Sh 900–2,560 (US$10–25) per month. A Rise in Self Interest Over the Community Good Young men used to be the community’s warriors, bestowed with the responsibility to protect and acquire property for the community through mass raiding. Young men usually acquired prestige by being brave and successful in predatory raids and by accumulating large herds of animals. With the coming of a cash economy, young men engage in violent raids for the sole purpose of selling the livestock to benefit themselves. Lessening Respect for Elders To fulfill individual responsibilities in accordance with customary law, each member of the community had roles to perform—depending on age, sex, and group affiliation—that maintained social organization and supported cultural norms and values (Carlson et al. 2012). Young men relied heavily on the decision making, knowledge, and governance of the elders regarding systems of animal husbandry. As pastoral livelihoods dimin- ished, the elders became less central to many young men who started engaging with the new livelihoods that emerged to fill this gap. The youth are now operating beyond the 136 Appendixes management capacities of the elders, also decreasing their relevance in terms of cus- tomary authority. This presents a major challenge to the customary institutions and the social stability these institutions were meant to protect. Many elders unable to resolve this growing crisis are increasingly turning to state justice and security providers to manage the growing rift between male youth and elders (Carlson et al. 2012). A classic example is cattle raiding, which was traditionally sanctioned by the elders but which must now be left to the state authority to regulate. In some cases, elders order fines for reparations or transgressions to be paid in cash rather than in animals. Youth are frustrated with the elders’ inability to manage the economic well-being of their com- munities. They are disdainful of the merits of customary authority when their elders lack the power to strengthen the well-being of the community (Carlson et al. 2012). Change from Community Defenders to Community Attackers The socialization of Turkana youth to be brave, to raid, and to protect the community creates a lot of hostility in them. Many youth (especially young men) who are unem- ployed or underemployed resort to theft and armed raiding to deal with their poverty and hunger, thus becoming a source of insecurity in Turkana County, especially in the refugee camp. What New Opportunities Portend for Youth In Turkana West Devolution Given the growing political interest and awareness of rights, young men have orga- nized community-level youth groups to heighten their political participation locally while also increasing their efforts—peace-building or development—within their com- munities and across groups. Constitutional provisions, particularly devolution, have increased the involvement of youth in formal political processes, thus allowing them to increasingly exercise their autonomy apart from the elders’ authority. This is indicative of the interest and agency among male youth to move beyond the traditional power relations with elders (Carlson et al. 2012). The young men are also coming together to support their own needs for livelihood opportunities. The ethnographic data findings of the worry analysis for the Turkana of Kakuma confirm that men, especially those who are young and middle-aged, are concerned with formal employment and related issues, such as illness, medical facilities, and disabilities, while women are more concerned with issues of hunger and, in the case of middle-aged women—the most consistent pro- viders of income and food for the family—about their lack of money to pay school fees. Because middle-aged women have established a portfolio of diverse livelihood activities E. Background Paper: 137 The Future of Kakuma’s Youth—Opportunities and Challenges ranging from domestic work in the camp to agriculture, collecting firewood and char- coal, fence-making, and construction, they experience the lowest level of worries over poverty and a hard life, even as they perform the bulk of labor in the camp and town. Older men also exhibit a lower rate of worries about poverty compared with others, probably because they either have a higher status and access to livestock or because poverty is normative and therefore not expressed as a worry or concern. In sum, the ethnographic and worry analyses, disaggregated by age and gender, reveals a variation in the general underlying concerns and tensions faced by the Turkana on a daily basis while they try to build lives in Kakuma, even as they adapt to the changing demograph- ics and issues brought by the refugees and the relief and development organizations. Discovery of Oil There has been excitement around the emergent oil industry and what it portends for youth in the region even though the country is at the preliminary stages of exploration and therefore the industry has so far only offered a small amount of material benefit to community members. Many youth look forward to being employed in the sector, particularly young men, who see the sector as a promising opportunity to gain wage labor and accumulate wealth in order to purchase material goods such as cell phones and radios (Enns and Bersaglio 2015). The sector was seen by older women as a viable way to “give youth hope,” meaning that the sector will help improve the quality of life among those who get jobs in the sector. While most youth in Turkana are not very well educated, they still express hope that they will get high-ranking, formal technical employment in the extractive sector, particularly because these companies provide training opportunities. Because they hail from the area where the resources will be mined, the youth hope to be given preferential treatment. The reality, however, as illustrated in table E.6, is that most director and management positions are occupied by expatriates. The few nationals in management are mainly surveyors, advisors, and engineers. “Skilled” employees are usually nationals, includ- ing welders, plumbers, electricians, crane operators, assistant drillers, masons, group financial controllers, ecologists, and sociologists—all mainly from other parts of Kenya. Unskilled workers are all nationals and are predominantly from the local community. Because the quality of employment matters in establishing whether the type of employ- ment activity that an individual is engaged in will drive them out of poverty, it is highly unlikely that the employment opportunities provided by the extractive operations in Turkana will provide a large push out of poverty for residents. The resulting effects on youth are discussed below. 138 Appendixes Table E.6. Employment and Gender Dynamics of the Employment Initiatives N=2149 Expatriates National Female Male Director 66.7 33.3 0 100.0 Management 69.7 30.3 11.4 88.6 Skilled 3.4 96.6 6.2 93.8 Semiskilled 25.3 74.7 3.8 96.2 Unskilled 2.4 97.6 13.1 86.9 Average (percent) 17.5 82.5 9.4 90.6 Source: IEA 2014. Reinforced Narratives of Mathharau The lack of permanent, unskilled manual jobs available to Turkana men within Turkana compounds the larger issue that many of these menial jobs have the potential for inter- actions of indifference or mistreatment that create experiences and reinforce narratives of mathharau and ng’imeny. Labor Migration and High Unemployment The existence of oil will trigger labor migration to exploration sites as well as high unemployment because many young people believe that being closer to sites of oil exploration will increase their chances of being directly employed by an oil company, especially for tasks requiring unskilled labor. The exploration sites are also thought to provide opportunities for entrepreneurial activities and better access to markets for products sold by youth. These mining areas are growing at shocking rates. According to the Economist, as cited by Enns and Bersaglio (2015), some towns have seen a popula- tion growth of 500 percent. Numerous studies have proved that the growth of urban centers with high unemployment rates (because of low education levels), no social ser- vices, and a young demographic will certainly have profoundly negative consequences on the development and stability of the region as a whole. Idle Youth Caught Up in Various Vices Because these youth have no relevant skills, education, or assets to compete for employ- ment opportunities, the cycle is further reinforced by groups of Turkana youth, mainly young men in the towns “sitting idle by the roadside chewing mirah, smoking cigarettes and shisha, and even consuming alcohol and drugs.” The emergent narratives among external actors, both relief and development workers and refugees (in Kakuma) is that of Turkana men being “lazy” and in general “unable to be productive.” E. Background Paper: 139 The Future of Kakuma’s Youth—Opportunities and Challenges Social Stratification As evidenced by the ethnographic data, social differentiation is also felt in relation to access to jobs. Class, contacts, and the ability to access cash posed major barriers for all the young men and women trying to secure these jobs. One man said: “Most businesses, they only hire if they know you or your family for the good jobs. For the menial jobs, the Turkana are cheaper. But incentive staff, that is something else. It does not pay as much, the Kenyans [non-Turkana] get much more for doing the same or even less work. But it is work. For that, you have to take courses, and sometimes, you have to pay hongo [bribe] to the person who is hiring, so that they will pick you. Other times, some people who get the jobs, they get it through leaders in the camp or town, so they have contacts. So that becomes a problem because not everyone has the contacts or the money.” (KII 2015) Triggers of Conflict Social stratification has bred competition for opportunities and rivalries among Tur- kana youth but most intensely between Turkana and non-Turkana youth from other parts of Kenya. As young people become more aware of how limited and competitive the opportunities in the emergent oil industry are, past exclusion from formal development processes is unearthing old wounds, and Turkana youth are becoming quite vicious. There is a belligerent resentment of up-country populations who are seen as coming to take the few opportunities that exist for the Turkana in this time of plenty—because of the discovery of oil—and yet during their time of need, the Turkana were neglected; no one wanted anything to do with them. Although conflict has been a constant feature of life in Turkana, the existence of the extractive sector will change the nature of conflict (table E.7). Other conflicts revolve around the complexi- Table E.7. Perceptions about ties of land. A key element in mitigating vul- Causes of Violence in Turkana nerability in this region has been the corporate Turkana use of land meant for grazing. The ability of  Cause of Violence (percent) elders to negotiate access with other groups Boundary/land disputes 9.0 as part of a long-term strategy has been key Politics/leadership disputes 0 to surviving droughts. Given that rainfall is Problems over livestock 72.7 sporadic and that resource availability con- Ethnic/tribal differences 9.2 sequently varies over time and space, there Source: IEA 2014. 140 Appendixes has been a framework for groups to avoid zero-sum competition and instead focus on mutually beneficial land-sharing schemes (Carlson et al. 2012). Unfortunately, the discovery of oil and gas is already interfering with this system and contributing to the breakdown of these cooperative practices. In places where there is the potential for oil deposits, communities are being displaced without proper procedure or compensation mechanisms. Restricted access to grazing land and water resources at oil excavation sites are cause for concern among pastoral communities. The sudden fragmentation of land and new restrictions on mobility and resources are destructive to pastoral livelihoods, which depend on the ability to move and access resources in response to environmental stressors (Enns and Bersaglio 2015). Changing land-use patterns in Turkana are further complicated by the fact that it is extremely difficult to compensate for the loss of mobility and migratory routes. There is also no way of valuing community land, and therefore the potential for oil exploration efforts to undermine pastoral land use is leading some to question the compatibility of oil development and pastoral activities in Turkana (Enns and Bersaglio 2015). These issues have greatly contributed to unrest, mistrust, and conflict between groups, mainly involving youth. Concerns have also been raised around environmental issues, including the contami- nation of community resources such as water and grazing land due to oil spills. These risks have been cited as further challenges to the well-being and survival of pastoral households in Turkana. According to Enns and Bersaglio (2015), despite these worries, not a single youth they interviewed suggested that oil development should be stopped altogether. Gender Discrimination There is an inherent lack of opportunities for women in the oil mining industry as evi- denced in table E.6: only 9.4 percent of all employees in this sector are female. This is a concern for young women already facing cultural marginalization that makes it difficult to successfully diversify their livelihoods. E. Background Paper: 141 The Future of Kakuma’s Youth—Opportunities and Challenges Conclusion There is evidence that the Kakuma refugee camp has offered opportunities to youth in Turkana West in terms of introducing alternative livelihoods and increasing the stan- dard of living in the area through increased household incomes, which according to the ethnographic data has improved the nutritional status of the people in Turkana West more dramatically than those from other parts of Turkana. It can therefore be concluded that, at least in this regard, the presence of refugees has had a net positive impact on Turkana. However, the youth remain in limbo because the camp’s existence forces them to grapple with a set of new dynamics. Some of the key challenges are: n For years, the government has neglected its responsibility to promote develop- ment in Turkana. Because of this, while on one hand, the education sector is lauded as the solution to poverty among the local people; on the other, it is stifled by a lack of resources to support the necessary infrastructural development required to improve quality and completion rates. Access to education is minimal for most Turkana youth, resulting in high rates of unemployment, and youth who attend school do not learn traditional pastoral skills, creating a double-edged sword that cripples youth who end up not educated enough for professional careers and not skilled enough to succeed at pastoralism. This is why they are referred to as the “lost generation.” n Changes in social and economic systems have introduced complex dynamics that require various groups of people in Turkana West to change in order to accom- modate new realities. Women, for example, have had to support their households through resource exploitation, menial jobs, and petty trade. However, while wel- comed by young women, this economic transition from traditional livelihoods puts pressure on them to work “double shifts,” and while these new opportunities may generate more household incomes or better livelihoods, they have also been a source of heightened vulnerability for women in terms of gender-based violence. n New opportunities are changing power relations. Opportunities such as devolu- tion are introducing new forms of power relations that operate outside traditional norms. And the discovery of oil deposits in Turkana has unearthed old wounds of exclusion from the formal development processes, stirring up resentment in up- country populations. Turkana youth are exhibiting signs of growing belligerence. 142 Appendixes References Behnke, R. 2008. “The Economic Contribution to Pastoralism: Case Studies from the Horn of Africa and Southern Africa.” Nomadic Peoples 12(1): 45–79. Betti, M. 2010. “The Children of Eve: Change and Socialization Among Sedentarized Turkana Children and Youth.” Master’s Thesis, Department of Social Anthropology, University of Bergen. http://hdl. handle.net/1956/7382. Carlson, K., K. Proctor, E. Stites, and D. Akabwai 2012. Tradition in Transition: Customary Authority in Karamoja, Uganda. Somerville, MA: Feinstein International Center, Tufts University. http://fic.tufts. edu/assets/Tradition-in-Transition.pdf. CBS (Central Bureau of Statistics) and ILRI (International Livestock Research Institute). 1999. Geographic Dimensions of Well-being in Kenya; Where Are the Poor? Nairobi, Kenya: CBS/ILRI. CRA (Commission on Resource Allocation). 2011. “Kenya County Fact Sheets.” Government of Kenya, Nairobi. http://siteresources.worldbank.org/INTAFRICA/Resources/257994-1335471959878/Kenya_ County_Fact_Sheets_Dec2011.pdf. Davis, J. 2007. “Total Economic Valuation of Kenyan Pastoralism.” WISP. http://cmsdata.iucn.org/down- loads/kenya_tev.pdf. Eaton, D. 2008. Violence, Revenge and the History of Cattle Raiding Along the Kenya-Uganda Border. Hali- fax: Dalhousie University. Ebei P., G. Oba, and T. Akuja T. 2007. “Long-term Impacts of Droughts on Pastoral Production and Trends in Poverty in North-western Kenya: An Evaluation of 14-year Drought Early Warning Data Series.” In Droughts: Causes, Effects and Predictions, edited by J. M. Sanchez, 103–138. Hauppauge, NY: Nova Science Publishers. Ekaya, W. N. 1998. “Primary Productivity, Energy Flow and Nitrogen Cycling in a Rangeland Ecosystem.” PhD thesis, University of Nairobi. Enns, C., and B. Bersaglio. 2015. Pastoralism in the Time of Oil: Youth Perspectives on the Oil Industry and the Future of Pastoralism in Turkana, Kenya. The Extractive Industries and Society 3(1): 1–11. GOK (Government of Kenya). 2002. “Turkana District Development Plan 2002–2008.” Government Printers, Nairobi. GOK, RNG, and ROG (Government of Kenya, Royal Netherlands Government, and Republic of Germany. 1997. “Land Degradation Assessment and Mapping in Kenya.” Government Printers, Nairobi. Gulliver, P. H. n.d. The Turkana Age Organization. Harvard University. Accessed November 2, 2015 at http://onlinelibrary.wiley.com/store/10.1525/aa.1958.60.5.02a00100/asset/aa.1958.60.5.02a00100.pdf ?v=1&t=iufbr5ta&s=0ce4d3b5e86f4d01f27acc9b00db341d8754003f. E. Background Paper: 143 The Future of Kakuma’s Youth—Opportunities and Challenges Huisman, C. S. 2011. “Once Patriarchs and Warriors: Masculinity and Modernity in Karamoja, Uganda.” The Fletcher Journal of Human Security, 60–80. http://fletcher.tufts.edu/~/media/Fletcher/Microsites/ praxis/xxvi/CarrieStefanskyHuisman.pdf. IDRC (International Development Research Centre), Adaptation, and DFID (Department for Inter- national Development). n.d. Climate Change Adaptation in Africa Program 2009–10 Annual Report. http://www.idrc.ca/EN/Documents/reducing-vulnerability-among-pastoralists-in-northerns- Kenya.pdf. IEA (Institute of Economic Affairs). 2014. A Primer to the Emerging Extractive Sector in Kenya: Resource Bliss, Dilemma or Curse. Nairobi, Kenya: IEA. ILRI (International Livestock Research Institute) and VSF-B (Veterinaires Sans Frontiers–Belgium). n.d. “Livestock Development and Climate Change in Turkana District, Kenya.” ILRI, Nairobi, Kenya. https://cgspace.cgiar.org/bitstream/handle/10568/575/Turkana_discussionPaper_7.pdf?sequence=1. Kagwanja, P. M. 2000. “Refugee Camps or Cities? The Socio-economic Dynamics of the Dadaab and Kakuma Camps in Northern Kenya.” Journal of Refugee Studies 13(2). http://www.humanitarianin- novation.com/uploads/7/3/4/7/7347321/demontclos_2000.pdf. Kamau, W., n.d. Customary Law and Women’s Rights in Kenya. http://theequalityeffect.org/wp-content/ uploads/2014/12/CustomaryLawAndWomensRightsInKenya.pdf. KNBS (Kenya National Bureau of Statistics) and SID (Society for International Development). 2013. From Pulling Apart to Pooling Together: Understanding Kenya’s Inequality for Equitable Development. Nairobi, Kenya: KNBS. Krätli, S., and J. Swift. 2003. Understanding and Managing Pastoral Conflict in Kenya. Sussex: University of Sussex. Kurimoto, E. 2002. “An Arena Called Refugee Camps: A Report from the Kakuma Camp.” Africa Report 35: 34–38. Migosi, J., D. Nanok, C. Ombuki, and J. Metet. 2012. “Trends in Primary School Dropout and Comple- tion Rates in the Pastoralist Turkana County, Kenya.” Universal Journal of Education and General Studies 1(10): 331–38. http://universalresearchjournals.org/ujegs/pdf/2012/November/Migosi%20et%20al.pdf. Ogindo, H. O., et al. 2009. “An Assessment of Drought Induced Vulnerability of the Turkana Pastoral- ist Community Livelihoods in Northern Kenya and Its Ability to Cope with Climate Change.” http://webcache.googleusercontent.com/search?q=cache:hRKRa4lfo8oJ:www.waternetonline. ihe.nl/symposium/10/full%2520papers/Water%2520and%2520Land/Ogindo%2520H%2520O. doc+&cd=2&hl=en&ct=clnk. 144 Appendixes Ohta, I. 2005. “Coexisting with Cultural ‘Others’: Social Relationships between the Turkana and the Refu- gees at Kakuma, Northwest Kenya. Pastoralists and Their Neighbors in Asia and Africa 69: 227–239 69. http://www.researchgate.net/publication/237630636. ———. 2007. Marriage and Bridewealth Negotiations among the Turkana in Northwestern Kenya. Koyoto, Japan: Center for African Area Studies, Kyoto University. http://jambo.africa.kyoto-u.ac.jp/kiroku/ asm_suppl/abstracts/pdf/ASM_s37/Part1.pdf. Okoti, M., J. C. Ng’ethe, W. N. Ekaya, and D. M. Mbuvi. 2004. “Land Use, Ecology, and Socio-economic Changes in a Pastoral Production System.” Journal of Human Ecology 16(2): 83–89. http://www.kre- publishers.com/02-Journals/JHE/JHE-16-0-000-000-2004-Web/JHE-16-2-075-150-2004-Abst-PDF/ JHE-16-2-083-089-2004-Okoti-M/JHE-16-2-083-089-2004-Okoti-M.pdf. Putnam, R. D. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, N.J.: Princeton University Press. Schilling, J., F. E. O. Opiyo, and J. Scheffran 2012. “Raiding Pastoral Livelihoods: Motives and Effects of Violent Conflict in North-Western Kenya.” Pastoralism: Research, Policy and Practice 2: 25. http:// www.pastoralismjournal.com/content/2/1/25. TUPADO. 2011. “Turkana Pastoralist Organisation Incident Register 2000–2010.” Unpublished study. TUPADO, Lodwar, . UNHCR (United Nations High Commissioner for Refugees). 2016. “Camp Population Statistics by Coun- try of Origin, Sex and Age Group.” UNHCR. https://data.unhcr.org/horn-of-africa/documents.php?p age=1&view=grid&Country%5B%5D=110. World Bank. 2001. Educating Nomadic Herders Out of Poverty? Culture, Education and Pastoral Livelihood in Turkana and Karamoja. Washington, DC: World Bank. http://www.fao.org/fileadmin/user_upload/ drought/docs/Educating%20Nomadic%20Herders%20-%20Karamoja%20and%20Turkana.pdf. Background Paper: F Gender Dynamics in Turkana County and Kakuma Refugee Camp Rationale and Methodology Unpacking critical gender-differentiated norms, dynamics, and constraints within and between refugee and host populations is critical. Understanding disparities and factors of vulnerability that perpetuate inequalities and impede poverty reduction and economic growth could ultimately identify enabling pathways and interventions to improve asset bases and promote resilience across and between populations of concern. The development challenges in Turkana are immense. Poverty and deprivation is widespread, and education and health indicators remain among the lowest in Kenya, particularly for women and girls. Despite statutory provisions protecting against dis- crimination on the basis of sex, enforcement of human rights is a challenge, especially in light of customary practices that predominantly favor men over women and persis- tent cultural norms such as bridewealth and early marriage. Economic barriers for men and women are significant, and without targeted attention, they threaten to undermine efforts toward inclusive growth and broader regional stability. Gender-based violence is a widespread and regular experience for Turkana and refugee women and girls, but there is also evidence that men are being victimized. Child labor is evident, and men and boys are subjected to various forms of targeted violence. Appropriate consideration of key gender dynamics and relationships that influence or feed into pervasive disparities across groups in Turkana County and the Kakuma Refugee Camp is critical. In this context, this study examines gender dynamics across multiple dimensions including: (1) transformations in gender-differentiated roles and responsibilities in light of a range of stressors to household resilience and vulnerability; (2) gender dimensions to protection challenges, including varying forms of gender- based violence that confront both refugee and host populations; and (3) gender dynam- ics tied to masculinity and male identity—notably, the decreasing ability of Turkana and refugee men and male youth to realize exacting social expectations of manhood. 145 146 Appendixes In the context of this report, gender refers to socially constructed roles and relation- ships between women and men. These roles are influenced by variables such as age, race, class, and ethnicity. They are mutable over time, subject to learning and evolving contexts. Diagnosis and analysis of gender-based differences has important implica- tions for poverty reduction and sustainable development, particularly in the context of persistent instability. Identifying and understanding the unique barriers and opportu- nities that men, women, boys, and girls each confront helps ensure that interventions, project activities, and analytical work promote the equitable realization of economic, political, and social gains. This analysis builds on an extensive literature review, complemented by an analy- sis of data on social and gender issues emerging within and between Turkana and refugee populations. Study methodology includes semistructured, in-depth, quali- tative interviews and focus group discussions conducted with local host community members. It is important to note, however, that this paper draws from interviews with a select sample of stakeholders rather than from the full cohort of interviews conducted during the course of the social impact analysis.8 Background Turkana County, the largest county in Kenya, is located in the northwest corner of the country and is home to over 1.2 million people. The environment is character- ized by a harsh arid and semiarid climate, limited access to water, and low agricultural potential. Turkana is among the five poorest counties in Kenya, reflective of a legacy of marginalization, underdevelopment, and neglect. Socioeconomic indicators are dire, with 87.5 percent of the population living below the poverty line (KNBS 2014). Poverty in Turkana is exacerbated by harsh environmental conditions, limited infrastructure, and minimal access to basic services (Opiyo et al. 2014). Turkana rates as one of the most deprived counties in Kenya across at least seven of the following variables: pov- erty, mean household expenditure, education, work for pay, water, sanitation, cooking fuel, access to electricity, and improved housing (KNBS and SID 2013). Only 39 percent of the population has access to improved water sources, and only 9 percent has access to improved sanitation. There is a limited differentiation in access between male- and female-headed households (KNBS and SID 2013). One report sug- gests that on average, 75 percent of households walk 2–3 kilometers to access sources of safe water (Wawire 2003). Malnutrition and undernourishment indicators are among 8. Additional data mining and analysis of available interviews may reveal further dynamics not reflected in this paper. F. Background Paper: 147 Gender Dynamics in Turkana County and Kakuma Refugee Camp the worst in the country; 23 percent of children are considered wasted,9 and 34 per- cent are underweight for their age. National averages are 4 and 11 percent, respectively (KNBS 2014). At an average of 6.9, the total fertility rate is three points higher than the national average of 3.9, and the county is growing quickly, with over 1.2 million people living in the region and an expanding population, particularly among youth (KNBS and SID 2013). Over 54 percent of families have more than seven children (Opiyo 2014), and almost half of the population is under 14 years of age. Higher fertility rates correlate with living in rural areas, deprivation, and low levels of educational attainment, and they further align with a low prevalence of contraceptives (10 percent). Only 15 per- cent of the population has achieved a primary-level education, and only 3 percent have attended secondary school. Turkana County demonstrates some of the poorest health indicators in the country, compounded by a limited investment in staffing, health sys- tems, and infrastructure, and a largely mobile pastoralist population. Maternal mortal- ity in the region is three times higher than the national average (Human Rights Watch 2015). Only 23 percent of women in the county were attended by a skilled provider during delivery, and only 23 percent delivered in a health facility (KNBS 2014). Traditional patriarchal practices persist, include polygamy, wife inheritance, and early or forced marriage. This is evidenced even among respondents of qualitative interviews: several of the interviewed women report having been married by the age of 12 or 13. Men are ascribed the primary responsibility of protecting and providing for the home while women attend to a range of domestic tasks in addition to activi- ties to supplement livelihoods and household income. These divisions, however, have evolved out of historical dynamics and continue to change as they are challenged and transformed in the face of evolving external stressors. The majority of households (55 percent) practice pastoralism, almost 16 percent rely on agropastoralism, 8 percent on fishing, and just over 8 percent on urban or periurban livelihoods (Opiyo et al. 2014). Households in Turkana confront a range of challenges related to the impacts of climate change, with vulnerability extending from diverse factors such as widespread poverty, food insecurity, recurrent droughts, land degradation, inequitable land distribution, and overdependence on rain-fed agriculture (IPCC 2012). 9. Wasting describes the fall of a child’s weight-for-height below minus two standard deviations, reflecting a failure to receive adequate nutrition. Wasting is often linked to illnesses such as diarrhea and deterioration in food supplies (KNBS 2014). 148 Appendixes Within this context of strained resources, poor infrastructure, and limited basic services, Kakuma refugee camp was established in 1992 with the arrival of a cohort of Sudanese young men commonly known as the “lost boys.” Kakuma refugee camp currently host over 160,000 refugees from 20 different countries (UNHCR 2016). South Sudanese refugees account for half of the population, with nearly 91,000 people reg- istered; Somalis account for 30 percent (just over 55,000); and Sudanese, particularly from Darfur, account for the next largest group, numbering just over 10,000.10 There are additional refugees from a wide range of countries, including Burundi, the Democratic Republic of Congo, Ethiopia, Eritrea, and Uganda. Unlike many other displacement settings where women and children account for the disproportionate share of the forc- ibly displaced, at Kakuma, there is a larger share (54 percent) of men and boys across almost all age ranges. The only exception is for individuals over age 60, of which women account for nearly 70 percent.11 As in Turkana County, deprivation and human insecurity prevail as refugees have poor access to most basic resources. Encampment policies mandated by the govern- ment of Kenya restrict the movement of refugees outside the camp, their ownership of livestock, and their engagement in formal employment and income-generating activities. Refugees are allowed to engage “voluntarily” in wage-paying activities, for which they are paid low-level “incentive” fees. While a vibrant informal economy has developed within Kakuma (Oka 2011, 2014) in which both the Turkana and refugees participate, encampment policies have been criticized for rendering refugees depen- dent on humanitarian support by external actors and for exacerbating divisions with local populations. While basic needs in Kakuma are significant, the administration of the camp by the United Nations High Commissioner for Refugees (UNHCR) and a range of other development actors ensures the provisioning of some critical basic services, includ- ing education, health facilities, water and sanitation, and other resources. Insecurity remains a challenge, however, with refugees confronting domestic and community violence, violence within and between national refugee groups, and sporadic violence between refugee groups and host populations. Select interviews with refugee popula- tions reveal mixed feelings regarding the Turkana, with several respondents reporting individual experiences of physical and sexual assault, theft, robbery, and murder. The Turkana report similar incidents by the refugee population. At the same time, some refugees identified hostile dynamics with other refugees as more likely, particularly the Sudanese. The incidence of violence seems to vary by year. Some respondents claimed 10. See the Kenya Kakuma Camp Population Data (2016) available on the UNHCR data portal at http://data.unhcr.org/horn- of-africa/region.php?id=12&country=110. 11. Ibid. F. Background Paper: 149 Gender Dynamics in Turkana County and Kakuma Refugee Camp that tensions have calmed recently and that the more serious incidents occurred in years past. For example, two Ethiopian respondents noted heightened tensions between refugee and Turkana populations in 2008, highlighting the outbreak demonstrations by refugee community leaders demanding improved protection from Turkana raiding. Gender Dynamics Among Populations Pastoralist populations in Turkana are deeply patriarchal, upholding distinct and differentiated roles for men and women in the household. A Turkana homestead, or awi, is traditionally comprised of a man; his wives and children; and potentially other relatives, such as his mother or other female dependents.12 Households are further orga- nized into an adakar, or herding group, comprised of a loose confederation of families who live and migrate together, thereby benefiting from collective protection from ban- dits or other raiding groups. Families are primarily patrilocal, with wives moving into the husband’s awi upon marriage. Current gender norms, which are relatively recent but now constantly reinforced, underscore the primacy of men in the household hier- archy, with women ascribed secondary and circumscribed status. As one respondent describes: “With us, in our culture and tradition, women were treated as kids … back then, their job was to be in the kitchen, to be seen, to be sent to forced marriage. It was common to have girls of 10 married to men of 50 or more.” (KII, 2015) There is evidence, however, that these norms reflect a relatively recent legacy of the presence of the British military, and that women previously occupied more promi- nent roles in household subsistence activities. Changes in social organization and settlement patterns emerged in the early 20th century, extending in part from events of severe livestock depletion compounded by skirmishes with the British military. In par- ticular, British forces rounded up women and children and confined them into enforced settlement groups of 10–15 awi households. Under these conditions, women and chil- dren were forced to remain in homesteads while men were removed from homes, either through forced labor, eviction, or forced recruitment into the army or police. These artificial divisions may have contributed to shifts in household roles and responsibili- ties. Turkana origin stories suggest that women previously occupied more prominent positions maintaining household support through hunting, foraging, and agriculture, but their forcible separation under British military rule may have contributed to the 12. However, this research, as well as another study, noted that at least initially, it is often difficult for a young man to set up his own independent awi due to labor demands for managing livestock herds or if he is unable to afford dowry, which can range from 50–300 goats (Ohta 2007). 150 Appendixes evolving dynamics and more distinct divisions of labor, with women becoming pre- dominantly responsible for the home (Vemuru et al. 2016). Under current structural dynamics of gender, men maintain normative roles as providers and protectors, responsible for activities related to livestock herding, household supervision, and decision making, as well as to household and commu- nal security. Women are responsible for domestic tasks related to cleaning and food preparation, child rearing, and firewood and water collection. Women also manage complementary functions related to livestock production, including providing water to herds, preparing for migration, milking livestock, and caring for small animals. In addi- tion, women tend to small ruminants and other animals, including goats, sheep, and poultry (Wawire 2003). Collectively, men and women are responsible for transmitting cultural values and systems and engaging in other activities to socialize young boys and girls, respectively, into their adult roles (Wawire 2003). Exposure to numerous and more frequent exogenous threats necessitate shifts in livelihood strategies. Households in Turkana confront a range of hazards that serve to exacerbate or magnify conditions of vulnerability, including shocks related to climate variability such as drought and flash floods, outbreaks of disease that affect livestock, and intercommunal conflict. These hazards, particularly climate-related stresses, result in a range of negative impacts, including food insecurity, loss of access to water and pasture resources, disrupted land-use patterns, and conflicts between and within households (Opiyo et al. 2014). In responding to these risks, men and women employ adaptive strategies to manage and maintain household welfare and security, such as increasing mobility; splitting herds; redistributing or diversifying livestock herds; accessing social security networks, including those with family and friends; and increasing reliance on alternate livelihoods, including farming, fishing (for those near Lake Turkana), collect- ing wild food, petty trading, and distributing food relief (Omolo 2010). Engaging in casual labor such as cleaning, washing clothes, construction work, and fence mending is common, particularly among Turkana populations involved with the refugee economy living near Kakuma. Adaptive strategies adopted by the Turkana are facilitated by transitions in house- hold responsibilities, with women playing more significant roles in support of their households. Men still bear the primary responsibilities for coping activities associated with livestock, including herd splitting and migration, but otherwise, traditional liveli- hood options for men have shrunk significantly in recent years. As a result, women have become central actors in adaptive strategies and alternative livelihoods for the household. Depending on environmental conditions, external stressors may increase F. Background Paper: 151 Gender Dynamics in Turkana County and Kakuma Refugee Camp household reliance on agricultural production, which is predominantly carried out by women. Female respondents in the study note that technical knowledge about agricul- tural practices are primarily transferred by women to girls, although some respondents say that if their sons were interested, they would teach them as well (FGD and KII 2015). Several women note that they had taught other women in their community as well and highlight improved farming as a preferred area of intervention in the event that additional development support is provided (KII 2015).13 Women are extensively engaged in the collection of wild foods, the making and sell- ing of charcoal, the collection and selling of firewood, casual labor, and managing small enterprises. As noted during one focus group discussion, women have organized economic associations and cooperatives across a number of activities as a microfinance mechanism for small business or activities related to livestock exchanges (FGD 2015). Polygamy is a common practice with implications for adaptive livelihoods. Men with multiple wives are responsible for the allocation of resources across wives and their respective children. While polygamous families can demonstrate adaptive strategies during conditions of drought, including shared allocation of resources and labor across wives and children, these strategies are not necessarily sustainable over protracted peri- ods (Wawire 2003). Of the respondents interviewed, 10 reported living in polygamous households. Relationships with co-wives seem to vary. Some women report having supportive relationships in which each wife shares the responsibilities for caring for children and providing for the home. Others report definite divisions between family members, including co-wives living in different homes. Indeed, one study notes that the splitting of polygamous families is not uncommon, particularly during periods of external stress, such as drought, when resources are limited. Women sometimes relo- cate to their own homes during a drought. The study notes that tensions between wives may increase during seasons of drought (Wawire 2003). Assessing current trends and contributing factors requires more research. Engagement with the refuge economy has become increasingly critical to house- hold income, and women are the central economic and social interlocutors between refugee and host populations. Among the Turkana living close to the Kakuma Refugee Camp, interaction and engagement with refugee populations has facilitated a critical means of survival through economic interactions and improved social cohesion. Social interactions and personal ties are facilitated in part through economic activities and 13. The extent to which agricultural production presents a viable option for communities, however, depends largely on location, availability of water, and other environmental conditions. 152 Appendixes exchanges in the camp. Turkana men report regularly looking for jobs to perform within the camp, including small trade, construction, ditch digging, and other low-skill or menial labor for nongovernmental organizations (NGOs) or refugees. Turkana women engage with refugees in the trade of milk, firewood, and charcoal. They also perform a range of domestic tasks, including cleaning, washing clothes, cooking, mending fences, and fetching water. While engaging in these activities is crucial to earning an income to support the household, it requires regular interaction with refugees, sometimes lead- ing to deeper and lasting personal bonds. One study in particular notes the distinc- tive ability of women to foster relationships between the Turkana and refugees (Ohta 2005b). These relationships and networks often lead to increased support from refugees in times of need. Several of the interviewed Turkana explain that even when work is unavailable in the camp, friends among the refugees offer food and other resources to help the Turkana in times of need (KII 2015). Engagement with the refugee economy in particular can hold both positive and negative implications for Turkana men and women alike. As mentioned, refugees provide a critical source of income and economic support for Turkana populations, leading many to note their near total reliance on the presence of the camp for household survival. For women in particular, as they bear increasing responsibility for providing for the home, engagement in income-generating activities, however marginal, as a mat- ter of survival and support for the home, creates a space for women’s empowerment and agency. At the same time, many Turkana men and women alike note the risk of assault when asking for payment or attempting to negotiate fees for work. As one respondent describes: “Three months ago, one mama went to get 250 shillings that one family owed her. They beat her so badly, that she had to be carried home by some men. She died in two days of the beating”14 (KII 2015). For women, engagement in certain livelihood activities, including charcoal firewood collection, increases their exposure to the risk of violence, including sexual assault. Some particularly vulnerable women reportedly resort to sex work. Proximity to the camp exacerbates vulnerabilities among Turkana children and youth. Child labor and exploitation, particularly among the Turkana, has been reported as a significant challenge as children seek to earn incomes, primarily from odd jobs within the camp. Framed as abuse during interviews, numerous respondents noted that children were not in school but were instead seeking to supplement their household income through odd jobs for refugee populations. 14. Most respondents noted challenges with predominantly South Sudanese refugees, particularly with the Dinka and Nuer; several associated Somalis with an increased incidence of poisoning. F. Background Paper: 153 Gender Dynamics in Turkana County and Kakuma Refugee Camp “There is a high demand for child labor, so they get money. And they drop out of school. One boy told me ‘mwalimu, if you make me stay in school, how can I eat? I get only lunch in school, not dinner, not breakfast. If I don’t work in camp, how can I buy clothes, even uniform.’ Many young boys go to school and then come to work in the camp.” (KII 2015). Several interviews highlight the potential for children to become vulnerable to exploita- tion or recruitment into gangs or criminal activities by older youth; others highlight their increased exposure to violence, subjecting them to the risk of physical assault (KII 2015). Despite substantial contributions by women to the contemporary pastoral economy, decision-making authority and power continues to reside with men. Several survey respondents note that household decision making is shared by male and female mem- bers, suggesting that as women’s contributions to household economies increase so too might their participation in key household decisions. However, several studies note that men continue to control the use and allocation of economic resources, investment decisions, and migration plans (Omolo 2010). Women are still charged with managing domestic tasks and the broader household economy despite their active engagement in income-generating activities, making them bear multiple and time-intensive burdens related to the care of the household as well as economic survival. Household vulnerability to climate-induced stresses and other external shocks is also distinctly gendered, with women being the most vulnerable. A recent microlevel study of household vulnerability in the Turkana region revealed that the majority of highly vulnerable households were headed by women or someone with a limited or no primary education (Opiyo et al. 2014). According to this study, 75 percent of female- headed households, accounting for 50 percent of the survey population, are more likely to be vulnerable to climate-induced shocks, stresses, and variability than male-headed households. Women’s vulnerability is likely linked to discriminatory norms and atti- tudes that circumscribe a woman’s role in the community and in the home, limits her access to educational and employment opportunities, and inhibits customary rights to productive resources, including land, property, livestock, and other assets. A myriad of key determinants further influence vulnerability. They include level of education; age of head of household; size of household; number of dependents; marital status, with higher rates of vulnerability among divorced or widowed persons; social connections, influenced by amount of time spent in an area; and access to extension services and early warning information. These findings further highlight the intersec- tion between poverty and vulnerability, noting that the majority of households exposed 154 Appendixes to external stresses and shocks, such as droughts, will probably require some form of assistance in order to recover (Opiyo et al. 2014). In light of the increasing proportion of female-headed households in the arid and semiarid lands of Kenya due to death of a spouse, separation, or divorce, it is critical that resilience policies and programming specifically target these populations. Ethnographic data emerging from interviews conducted for this study highlight vulnerabilities confronting single and abandoned women, particularly widows. The death of a spouse or father leaves many women without legal, social, or economic pro- tection. Several informants relate experiences of being excluded from an awi along with their children. Women are granted limited access to or control over livestock herds, minimizing their opportunities to earn incomes. The worry analysis further relates deep concerns around widowhood that older women confront, with many seeking protec- tion as concubines or co-wives of their deceased husband’s brother. For women seeking protection by returning to their father’s awi, particularly in the case of divorce, social dictates sometimes require the return of any dowry paid by the husband’s family, which in many contexts serves as a deterrent to separation. In many instances, single or aban- doned women with children migrate to more urban environments, such as Kakuma, Lodwar, or Lokichoggio, in order to earn an income. These women and sometimes their children tend to become part of the larger low-income labor force, and interactions between female Turkana and refugee communities have reportedly increased. While not explicitly explored in the study, gender relations within refugee com- munities continue to reflect norms and dynamics of countries of origin. This study did not conduct ethnographic interviews to distill current gender dynamics within and across refugee communities, but there is evidence to suggest a replication of norms and relationships as transmitted from communities of origin. As in Kenya, refugees from South Sudan, Somalia, and other countries perpetuate uneven power dynamics and patriarchal patterns that contribute to persistent disparities between men and women and boys and girls. While humanitarian and development programming attempt to remediate disparities by paying particular attention to education and health program- ming, social and economic empowerment interventions, and peace-building initia- tives, women and girls continue to confront entrenched dynamics that minimize their autonomy, their voice, and their agency. As discussed in the proceeding section, female refugees frequently confront multiple vulnerabilities that expose them to a range of protection challenges with associated broader implications for their well-being. F. Background Paper: 155 Gender Dynamics in Turkana County and Kakuma Refugee Camp Gender-Based Violence and Protection Challenges Gender-based violence is a significant challenge in the region. It is often rooted in entrenched norms and practices that perpetuate uneven power dynamics between men and women. However, conditions of conflict and displacement—whether protracted or acute—further exacerbate incidents of gender-based violence as manifested in varying forms of physical, sexual, psychosocial, economic, and emotional abuse. Factors driving the persistence include the breakdown of the rule of law, including security and justice institutions, and the social and moral order that might otherwise have guarded against it. Gender-based violence, particularly sexual and physical assault, has further been used as a strategic tool of conflict, intended to disempower and humiliate opposing groups, particularly men unable to protect their families and homes. Indeed a global review of 50 countries highlighted significant increases in gender-based violence fol- lowing major wars (World Bank 2011). In the context of displacement, human insecu- rity prevails, particularly in camps comprising very small spaces with limited security measures or other mechanisms of protection. Vulnerability acts as a multiplier across multiple dimensions—social, economic, and legal—to the compounding effects of gen- der-based and structural violence that extend out of women’s broader social margin- alization and low social status. Psychosocial dislocation and trauma originating from acute conditions of flight as well as broader conditions of marginalization and exclusion often manifest themselves in harmful behaviors that can lead to aggression and violent behavior, particularly among men. The incidence of gender-based violence in Kenya is a significant challenge across regions and socioeconomic conditions. The 2009 Kenya Demographic and Health Survey reports that 45 percent of women aged 15–49 have experienced either physical or sexual violence, and 14 percent have experienced both. Nearly half (47 percent) of women who have ever been married have experienced emotional, physical, or sexual violence. The likelihood of becoming the victim of violence increases with age (KNBS 2009).15 In particular, intimate partner (domestic) violence, is prevalent: 49 percent of women report having experienced physical violence in the home (KNBS 2014). The magnitude of the challenge is likely to be even greater given the sociocultural norms that stigmatize survivors of violence, consequently contributing to underreporting and deterring health or justice-seeking behaviors. 15. While a key indicators report is available for the 2014 Demographic and Health Survey in Kenya, these findings do not include figures for forms of gender-based violence occurring outside the home. Thus, some figures included here reflect the 2008/09 survey report. 156 Appendixes Gender-based violence also constitutes a serious and prevalent protection challenge within Turkana County and in the Kakuma Refugee Camp. In-depth interviews and focus group discussions with respondents from Turkana reveal high and continuous exposure to varying forms of violence. There is a distinct intersection between liveli- hood insecurity and incidents of violence because women and girls engage in high-risk activities to earn incomes. They are particularly vulnerable to physical or sexual assault while collecting wood or other resources to support the home. Several female firewood collectors report a commonality of experience with regard to violence, especially rape. One woman discusses the money that could be earned mending fences, but notes that collecting raw materials for the process is dangerous, putting one at risk of being robbed, raped, or otherwise assaulted. Perpetrators are equally likely to include Turkana men and male refugees (KII 2015). The interviews provided insufficient information to link instances of rape with a particular tribe or ethnic group.16 Women have adopted mea- sures to protect against being assaulted, including traveling in large groups and carrying axes and pangas, but exposure to violence is practically ubiquitous. The impacts of assault can be multiple and severe. Women can contract HIV/AIDS and/or suffer from psychosocial distress. Stigma and shame deter reporting and health- seeking behaviors, leaving many to suffer alone or become more ill: “Many don’t even go to doctors, but they get sicker and sicker, the get depressed after the rape. They can- not tell anyone, they will get beaten and kicked out of their homes by their men. Many will even commit suicide.” (KII 2015). Stigmatization is further tied to rejection by both households and communities, which several individuals note could lead to even riskier behaviors for female survivors of violence, including resorting to commercial sex to earn an income. According to one man, “Because there is so much hunger and poverty, women have to feed their children, sometimes they are raped when collecting firewood, and they get scolded by the community, they are shunned. They then have to turn to pros- titution. The Turkana people don’t like that commerce but hunger speaks.” (KII 2015) Despite negative impacts on health and well-being, however, the economic imperative to earn an income and survive appears to overwhelm security concerns or fears around exposure to violence. As one woman put it: 16. Several studies, for example, have noted prevalence of gender-based violence perpetrated by Somali men against refugee women. F. Background Paper: 157 Gender Dynamics in Turkana County and Kakuma Refugee Camp “Please understand, that being raped here in Kakuma, is something that we all know, and we, some of us, have undergone at some point in our life. You can’t stop working just because something happens that is bad. You have to get the firewood, or the charcoal, or go back to camp to work for the refu- gees, because your children are home and they get hungry, they need to go to school. Maybe they can change things after they’re educated, but we need to do what we need to do. One bundle of firewood, if we sell it, even after a bad day, that is 100 shillings” (KII 2015). Interestingly, during the interviews, despite the potential for stigmatization and shame, for the most part, women were very open about their experiences of being assaulted. During one focus group discussion, women related the commonality of the experience and openly referenced their own experiences of violence.17 Women even laughed openly during the discussions of their experiences. This may be indicative of an adaptive coping mechanism to rationalize or normalize the experience (interviewer’s note); the psychosocial impact is likely severe. Distrust of state security institutions is high, which likely further contributes to underreporting. Nearly all respondents relate frustrations and resentment toward local and national leaders specifically and government structures more generally with regard to insecurity and violence. Turkana men and women reveal high levels of distrust and fear of the police. Respondents describe police as often dismissive of assault claims, blaming survivors for their experiences of violence, and several note that the police are themselves sometimes the perpetrators of physical and sexual assault. As one woman notes: “I feel safe in the camp, walking, working, even as a woman. But I feel scared of the police, they will not hesitate to beat you if you are [in the] camp late, if you are even 10 minutes late they can beat you badly.” (KII 2015) The Incidence of gender-based violence among refugees in Kakuma is common. Conditions of insecurity within the camp prevail, with a reported range of violations that include community, domestic, intimate partner, and sexual violence (Horn 2010; Mwangi 2012). Forced marriage, abduction, child labor, and sexual exploitation are reported as significant challenges. A recent needs assessment jointly conducted by the World Food Programme (WFP) and UNHCR reveal that female children are com- monly used as salespeople for businesses, as laborers, and as transactional sex workers 17. In keeping with ethical guidelines, women were never asked directly to describe their experiences of violence, but many raised the issue of their own volition. Men also raised the issue, reflecting its pervasiveness in the lives of many people. 158 Appendixes (WFP and UNHCR 2014). Studies highlight the vulnerability of newly arrived refugees, especially single females, to varying forms of violence, including rape, and the lack of access to basic needs increases the likelihood of a woman engaging in transactional sex for survival (Mwangi 2012). Reportedly, violations are common among South Suda- nese and Somali populations, but given tendency toward underreporting, the scope and extent of the challenge across groups is not known. Exposure to violence frequently results from efforts to fulfill domestic responsibili- ties and meet basic household needs. Like the women of Turkana, collecting firewood, water, or other key resources increases the refugee’s risk of exposure to violence, par- ticularly of sexual and physical assault. Incidents of rape have also been reported during the distribution of food rations. However, a recent study notes a disturbing shift in recent years from incidents of rape occurring while women were conducting domestic tasks far away from their homes to more direct attacks within homes (Mwangi 2012). There is anecdotal evidence of male experiences with sexual violence in Kakuma, but given the even more stringent social taboos around male rape, concrete data is virtually nonexistent (Horn 2010; Mwangi 2012). According to data from Lutheran World Fed- eration, out of 530 reported incidents of sexual and gender-based violence in 2011, 61 were reported by men (HRC and UNHCR 2013). As in other contexts, incidence data are generally limited, and the information that does exist is likely an underestimation due to low levels of reporting. Intimate partner violence is a significant problem among the Turkana and the Kakuma refugee communities. This research does not include reference to incidents or drivers of intimate partner/domestic violence in Turkana, although it constitutes a serious challenge in regions and socioeconomic settings across Kenya. As indicated earlier, nearly half of women who have ever been married have experienced some form of emotional, physical, or sexual violence perpetrated by their partners. Most incidents are current, that is, they were committed within 12 months prior to the administration of the survey (KNBS 2009). As in many countries, social norms that tolerate or justify violence against women in the home persist in Kenya, and women are socialized to accept and rationalize their experiences of violence by a spouse or relative. Women bear the health and psychological burdens of violence in the home, with roughly one-third reporting incidents of physical or sexual violence resulting in some form of physical injury (KNBS 2009). Conditions of displacement often give rise to elevated rates of intimate partner violence. Global experience in displacement settings indicate that intimate partner violence is often the most frequent and widespread form of violence experienced by affected populations. A systematic review of relevant literature examining gender-based F. Background Paper: 159 Gender Dynamics in Turkana County and Kakuma Refugee Camp violence in emergency settings finds that rates of intimate partner violence are extremely high across all studies—frequently higher than rates of wartime rape and sexual violence (Stark and Ager 2011). Another study notes that intimate partner violence escalates among displaced populations confronting restrictions on their mobility, inadequate access to livelihood opportunities, and limited access to authorities or decision-making processes. These challenges contribute to feelings of increasing frustration, dislocation, and disempowerment, which manifest in negative behaviors, including violence in the home. Consequently, intimate partner violence is recognized as the most common form of gender-based violence in camp settings (Horn 2010). Drivers of intimate partner violence in displacement settings are multiple and complex. A recent report from the International Rescue Committee analyzing con- ditions of displacement in Iraq, Kenya, and South Sudan demonstrates that intimate partner violence, including physical, emotional/psychological, and sexual violence, is one of most prevalent forms of violence confronting women and girls (IRC 2015). According to this study, key drivers contributing to the perpetration of intimate partner violence include: (1) rapidly shifting norms as women experience increased access to opportunities outside the home while men’s ability to do the same diminishes, increas- ing tensions in the home; (2) family separation, loss of social cohesion, and breakdown of community structures that might have provided systems of protection and support; (3) forced marriage, which may be sought as a mechanism for economic and social pro- tection but which often exposes women and girls to extreme dependence and dramati- cally imbalanced power relations—a significant challenge among the South Sudanese in Kakuma where abduction and forced marriage is common (Horn 2010); (4) conditions of extreme poverty, exacerbating tensions in the home; and (5) substance abuse as men access negative coping behaviors to address feelings of stress, depression, and frustra- tion that often emerge in the context of displacement (IRC 2015). A culture of silence around intimate partner violence and other forms of gender- based violence persist, driven in part by the associated stigma and shame and by sociocultural norms that contribute to levels of acceptance, particularly for violence in the home. This culture contributes to severe underreporting of gender-based vio- lence, masking a clear understanding of its prevalence. In the context of the study, female respondents spoke openly about sexual and physical assault experienced by women outside the home. However, no comments were shared about experiences of violence inside the home. A study in Kakuma found that most incidents of gender-based vio- lence, particularly intimate partner violence, were not reported to UNHCR, to police, to health providers, or to any other service provider (Horn 2010). In fact, respondents for this study expressed disapproval of women reporting directly to agencies and NGOs rather than working through existing community structures (Horn 2010). 160 Appendixes Within Kakuma, several agencies and organizations have attempted to respond to and reduce women’s exposure to gender-based violence. Core actors engaged in mitigation and response to gender-based violence include UNHCR, WFP, Jesuit Relief Service, and Lutheran World Federation, but survivors often prefer to work through community leaders and women’s support groups when seeking response or protection services (Horn 2010). These organizations have responded by providing response ser- vices to survivors, including medical and psychosocial support, and by establishing safe houses/havens for victims (HRC and UNHCR 2013). Mitigation and prevention programming seeks to address the issue of access to basic resources to reduce exposure to risks of violence. UNHCR, for example, adopted mitigation measures to reduce women’s exposure to violence through initia- tives addressing natural resource management and cooking fuels, albeit at a small scale (WFP and UNHCR 2014). Similarly, responding to protection challenges emerging from the collection of firewood for the home, WFP initiated a project in 2012 called Safe Access to Firewood and Alternative Energy (SAFE). The objectives of the project were to: (1) reduce refugee women’s risk exposure to violence through dissemination of fuel-efficient stoves and alternative fuels; (2) explore alternate energy technologies that could be applied to livelihood and protection needs; (3) promote the adoption of liveli- hoods that reduce women’s reliance on the collection of firewood for income; and (4) provide schools with fuel-efficient stoves to ensure that the cost of fuel does not impede school attendance (WFP and UNHCR 2014). The pilot in Kakuma was complemented by sensitization and awareness-raising activities around protection concerns and gen- der-based violence. Ironically, while these initiatives may have managed to reduce the incidence of violence among refugee populations to some degree, a Turkana woman noted that these initiatives have simply shifted victimization from the refugees to the Turkana because they are now the ones engaging in risky activities to bring services and resources like firewood to the refugees (KII 2015). Despite efforts to improve prevention and response mechanisms within the camp, the presence of these systems has not necessarily translated into greater protection for refugee women. One study notes that most survivors of violence remain unlikely to report incidents through formal channels (Horn 2010). Refugees are more likely to seek redress through community mechanisms that replicate or reflect customary institutions from their community of origin. Unfortunately, across communities, these structures emphasize reconciliation and community harmony, often at the expense of individual protection or rights. As such, these structures often conflict with international human rights principles, and determinations emerging from them can come at the expense of the protection of individuals or the provision of mechanisms for punishment or redress. F. Background Paper: 161 Gender Dynamics in Turkana County and Kakuma Refugee Camp Among Somali refugees, for example, customary systems are oriented around mecha- nisms of communal reconciliation as families seek redress through compensation (diya). An emphasis is placed on the preservation of social cohesion and pacification rather than on punishing perpetrators or protecting individual rights. Court determi- nations are often carried out by male tribal leaders or elders and thus likely reflect and uphold deeply entrenched patriarchal norms that ultimately disfavor women. Findings from numerous studies further confirm that women are still not receiving adequate protection from violence or responsive care, either because current mecha- nisms are insufficient or because women are unable or do not choose to access the available services (Horn 2010). Several studies have noted that efforts to provide protec- tion to survivors through safe spaces and shelter models are often misunderstood by the surrounding communities, which view them either as a punitive attempt to separate a survivor from her family or as a safe space provided for prostitutes (HRC and UNHCR 2013; Horn 2010). These misconceptions likely influence the extent to which survivors are able to access needed response services and underscore the critical need to improve community awareness and understanding of them. Several Turkana informants also noted experiences of marginalization and even corruption by relief actors, with reported instances of discrimination against the Turkana and extortion in exchange for access to services. These reports are of particular concern and if true may further contribute to conditions of structural violence experienced by the Turkana and refugees alike. Masculinities Conditions of deprivation and protracted displacement in Turkana County and Kakuma have likely contributed to an erosion of the markers of manhood and male identity. Male identity—like female identity—is relationally coproduced between men and women and between communities and the broader ecology in which individuals operate. One study notes the range of masculinities in sub-Saharan Africa, emphasizing that versions of manhood are: “(i) socially constructed; (ii) fluid over time and in differ- ent settings; and (iii) plural” (Barker and Ricardo 2005). At the same time, certain com- monalities prevail, particularly the primacy of the social requirement to earn an income and achieve financial independence and to then marry and have a family as central mark- ers of manhood (Ricardo and Barker 2005). Normative masculinities further designate men’s roles as fathers, protectors, and providers. In Kenya, male power is predominantly vested in patriarchal norms transmitted and reinforced through rites of passage, dowry or bridewealth payments, and virilocal residence (Amuyunzu-Nyamongo and Francis 2006). In the Turkana region, livestock (abundance or lack thereof) play a determinant role in a man’s status within his community and within the home. 162 Appendixes Since the 1980s, a range of socioeconomic, political, and cultural changes in Kenya have transformed gender relations, which has contributed to a reduction in the abil- ity of many men to achieve normative notions of manhood (Amuyunzu-Nyamongo and Francis 2006). Factors include depressed economic growth caused by corruption, mismanagement, declining global prices for cash crops, and other government failures; an expanding population; increasing vulnerability; unemployment; and declining access to basic services. Traditional livelihood systems in the Turkana region have eroded due to a range of adverse conditions, including drought, disease affecting livestock herds, desertification, and eroding natural resources, all of which impede pastoralist practices. As a consequence, men, but especially women, have diversified their livelihood activi- ties and engaged in adaptive coping strategies to provide for the home. Diminishing livelihoods in Turkana have contributed to the contraction in men’s economic opportunities. As mentioned, women are actively and significantly engaged in adaptive strategies and often bear a substantial share of the burden of both income- generating activities and attending to the home, making them seem better equipped to adapt to livelihood challenges. Conversely, men’s roles in livelihood and economic activities have contracted, reducing their contributions to the household economy and security. Relief activities targeting the needs of the most vulnerable are currently directed in part toward women, as evidenced by the Food-for-Assets program, which transfers cash grants primarily to vulnerable female-headed households. The proximity of the refugees and the attendant aid architecture available to address their needs high- lights the challenges confronting the Turkana and the absence of the state in addressing them (KII 2015). This accentuates perceptions among the Turkana of their margin- alization and neglect by the government as well as their feelings of resentment and powerlessness. Informant interviews emphasize a persistent narrative around notions of indig- nity—ng’imeny or matharau. Feelings of dependence, marginalization, and/or shame that arise from a reliance on refugee populations for an income, particularly in the face of emotional, psychological, or physical abuse, contribute to these experiences of ng’imeny, further compounded by perceptions of systemic exclusion by the state and other external actors. These notions emerge from historical and structural processes that are perceived to have begun with the arrival of outsiders to Turkana, whether for business, employment, development, or residence, and they are reinforced when Tur- kana go to other parts of Kenya. The broader narratives of distrust, anger, and exclu- sion that the Turkana have generated about the political, economic, and social system from which they have been marginalized for over a century shape their perceptions and beliefs regarding external development interventions. F. Background Paper: 163 Gender Dynamics in Turkana County and Kakuma Refugee Camp For refugee men, conditions of forced displacement can present significant chal- lenges to identity markers of manhood and male identity, often with negative impli- cations on broader social cohesion and stability. In the context of Kakuma Refugee Camp, restricted mobility combined with a lack of access to livestock or opportunities for otherwise earning an income contribute to men’s frustrations with life in the camp, and bring with them associated impacts on psychological and emotional well-being. As articulated in one study: “The condition of waiting and living at the mercy of the relief, repatriation and relocation process is a debilitating experience that leads to frustration, ennui, lethargy, self-victimization or a culture of victimhood, and in certain cases, to anger and outright violence.” (Oka 2011) Dislocation and disempowerment among men in displacement settings contribute to feelings of depression and aggression or even acts of violence, including an increased incidence of domestic or intimate partner violence as well as interethnic and intergen- erational conflict at the individual and group level (Horn 2010). Feelings of marginalization and disempowerment are further compounded by devel- opment actors, who often adopt gendered approaches to interventions—specifically targeting women and girls at the perceived exclusion of men and boys. Interventions advancing more gender-equitable norms and responses through distribution activities targeted at women or women’s empowerment programming can be interpreted by men as disempowering and as a further affront to their masculinity. The blurring or dis- solution of boundaries between men and women can further contribute to feelings of marginalization and resentment as well as to violence in the home as men attempt to reassert dominance and enforce more traditional male and female roles (Dolan 2003). Male youth in Turkana and in the Kakuma Refugee Camp confront significant chal- lenges. Normative associations between employment, marriage, and masculinity reflect important pathways, not just to manhood, but to adulthood. In the absence of oppor- tunities to work or earn an income, to get married or to achieve other markers of man- hood—such as receiving an education, owning land, or building a home—many men, particularly the young, perceive themselves as socially, culturally and/or, politically marginalized, and even entrapped as “youth” (Interpeace 2014).18 In Kakuma, refugees are not technically allowed to own livestock while among the Turkana, diminished live- 18. Like masculinity and male identity, notions of youth are ascribed to a range of identifying markers, not just age, such as gender, marital status, kinship networks, profession, educational attainment, class, and other variables (Interpeace 2014). 164 Appendixes stock assets impede payment of bridewealth requirements. Unable to pay a dowry or to fulfill other traditional requirements, many young men experience frustration at not being able to get married or to facilitate alternate transitions to manhood. Frustrated masculinities are manifest in a range of negative behaviors. Thwarted ambitions and disaffection are commonly channelled into negative coping behav- iors, including drug, alcohol, and khat addiction; aggression; violence; and crime. Inter- generational conflict and tensions are prevalent (Interpeace 2014). Although several Turkana respondents noted a general absence of enmity between Turkana and refugee populations, most acknowledged occasional incidents of violence between and across communities, especially recurring strife between the Turkana and the South Sudanese, including the Dinka and the Nuer. As one Turkana man noted: “Many people, many men, they drink a lot because the job is hard, and there is a lot of abuse, when you are insulted in the street. Many times, they don’t do that to us, because we are men and we can fight. But we see that happening to our women, to children, to old babas and mamas. We see young boys and girls chased by the refugees, chased away from their shops or homes. And it feels very bad. But most times, we can’t do anything” (KII 2015). Thwarted masculinities, particularly among Turkana youth, are probably connected to increased engagement in banditry and opportunistic cattle raiding in the region. Raiding activities among the Turkana have changed in recent years, partly influenced by conditions of deprivation and economic stagnation characteristic of the Horn of Africa region. Raiding, once an important cultural practice and mechanism for wealth redistribution, has shifted into a predatory activity where wealthy, commonly external actors pay male youth to engage in raids to acquire cattle intended for immediate sale rather than as a store of future wealth (Omolo 2010). The introduction and prevalence of small arms and light weapons has rendered the practice more lethal, and this has led to more frequent and lasting hostilities with neighboring tribes and increasing casual- ties involving women and children (Omolo 2010). The widespread inability among men to pay bridewealth has given rise to alternative marriage practices and informal partnerships, particularly between male refugees and Turkana women. Confronted by increasing deprivation and diminishing liveli- hoods, both Turkana and refugee men are increasingly unable to meet bridewealth requirements, which has partly contributed to shifts in marriage practices, especially the rise in informal marriages and elopements. While traditional practices have a space for these types of informal unions, particularly in Turkana culture, payment requirements F. Background Paper: 165 Gender Dynamics in Turkana County and Kakuma Refugee Camp persist when children are born, creating opportunities for tensions within and as interethnic marriages have emerged, between communities (Ohta 2005a). One study points out the potential for intergroup conflict between refugee and Turkana families emerging from a refusal or inability to pay the prescribed livestock payment upon the birth of a child (Ohta 2005b). However study respondents still emphasize the normative primacy of bridewealth among the Turkana and the impact of rising pressure on male youth to pay it (KII 2015). One informant noted a shift in marriage preferences with young women wanting to marry young rather than older men who might be better able to provide a dowry. This trend intensifies pressure on male youth and increases the potential for men to go into debt to enable marriage (KII 2105). Another study explores the emergence of the informal economy and its influence on household economic welfare as well as on the emotional or psychosocial dimen- sions of well-being (Oka 2011). In numerous interviews, refugees illustrate the extent to which informal commercial activities and increased consumption foster a sense of dignity or “normalcy” by enabling them access to and agency over their daily lives in the camp. Goods purchased through informal market networks enable wider social engagement, the establishment of social and kinship networks, and the opportunity for individuals to (re)construct social existence. As one Somali man described, “When I invited you to my house, I offered you tea, with milk, sugar, cloves, I bought cake, [sweetbread] as I should, you are my guest. I welcomed you as I would in Somalia. When I do this, I feel like a Somali man, I feel normal.” (Oka 2011) The study further noted that enhanced feelings of well-being and normalcy have con- tributed in some measure to the maintenance and sustainability of relief activities that otherwise fall short of addressing the needs and overarching stability of the camp. This finding underscores the critical need to investigate or pursue programming for both the Turkana and refugee populations to improve their economic welfare, their psychosocial well-being, and notions of dignity (arimatoi/heshima). Conclusion Men, women, boys, and girls in the refugee and Turkana communities alike con- front numerous challenges in their daily lives, which are further compounded by a range of external stressors, including poverty, climate change, environmental degradation, political marginalization, poor infrastructure, and limited access to basic services. The findings that emerge from this analysis emphasize the complexity 166 Appendixes of structural dynamics between men and women in Turkana, rooted in “traditional” notions of gender, which are actually relatively recent but which are continuously rein- forced. Although recent efforts in education outreach have transformed the attitudes of many men about legitimate pathways for earning an income or gaining social status, this study shows that maintaining herds, either within the household unit at Kakuma or with the larger awi, remains the primary pathway for incomes and livelihoods as well as communal notions of masculine identity. Livestock herds continue to be seen as central to enabling access to marital alliances through a dowry or bridewealth, as a contingency investment in the event of environmental or climatic shocks, and as a quick means to access cash for socioeconomic needs. The loss of livestock herds continues to detrimen- tally affect the economic resilience of communities as well as notions of ng’imeny and psychosocial dimensions of health and well-being. Alternate livelihood opportunities often lie in interactions with refugee communities, however, many of these opportuni- ties are informal, hard to obtain, and involve some form of graft or connections; and hence those without capital or connections may not find any access to subsistence opportunities. Reliance on refugees for an income further serves to entrench feelings of ng’imeny or matharau among the Turkana. Turkana women continue to occupy a circumscribed status in the awi-adakar sys- tem, and household vulnerability to external shocks is distinctly gendered, with women among the most vulnerable to abandonment, displacement, and other forms of struc- tural or physical violence. At the same time, diminishing livelihoods have necessitated women’s increased participation in and diversification of income-generating activities, particularly working in the refugee economy. Subsistence activities include a range of low-skill but sometimes intense labor, including collecting firewood, making charcoal, building fences, cleaning houses, washing dishes and clothes, and fetching food and water. Women participating in income-generating activities within and outside refugee communities in Kakuma are regularly exposed to experiences of physical and sexual violence as well as forms of emotional and psychological abuse. At the same time, women’s engagement with the refugee economy has become increasingly critical to household incomes, and women serve as the central economic and social interlocutors between refugee and host populations. Again, while a woman’s participation in income- generating activities can create a space for her empowerment and agency as she bears increasing responsibility for providing for the home, she also risks exposure to varying forms of gender-based violence that continue to be a prevalent challenge among both refugee and Turkana communities. F. Background Paper: 167 Gender Dynamics in Turkana County and Kakuma Refugee Camp References FGD = focus group discussion KII = key informant interview Amuyunzu-Nyamongo, M., and P. Francis. 2006. “Collapsing Livelihoods and the Crisis of Masculinity in Rural Kenya.” In The Other Half of Gender: Men’s Issues in Development, edited by I. Bannon and M. Correia, 219–244. Washington, DC: World Bank. Barker, G., and C. Ricardo. 2005. Young Men and the Construction of Masculinity in Sub-Saharan Africa: Implications for HIV/AIDS, Conflict, and Violence. Social Development Papers 26, Conflict Prevention and Reconstruction Series. Washington, DC: World Bank. Dolan, C. 2003. “Collapsing Masculinities and Weak States: A Case Study of Northern Uganda.” In Mascu- linities Matter!: Men, Gender and Development, edited by Frances Cleaver. London: Zed Books. Horn, R. 2010. “Responses to Intimate Partner Violence in Kakuma Refugee Camp: Refugee Interactions with Agency Systems.” Social Science and Medicine 70(1): 160–68. HRC (Human Rights Center) and UNHCR (United Nations High Commissioner for Refugees). 2013. Safe Haven: Sheltering Displaced Persons from Sexual and Gender-Based Violence. Case Study: Kenya. HRC and UNHCR. Interpeace. 2014. Rebuilding Trust: Annual Report 2014. Interpeace. IPCC (Intergovernmental Panel on Climate Change). 2012. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. Cambridge: Cambridge University Press. IRC (International Rescue Committee). 2015. Private Violence, Public Concern: Intimate Partner Violence in Humanitarian Settings. Practice Brief. New York: IRC. KNBS (Kenya National Bureau of Statistics). 2009. Kenya Demographic and Health Survey 2008–09. Nairobi, Kenya: KNBS. ———. 2014. Kenya Demographic and Health Survey 2014. Nairobi, Kenya: KNBS. KNBS and SID (Society for International Development). 2013. Exploring Kenya’s Inequality: Pulling Apart or Pulling Together. Nairobi, Kenya: KNBS. Mwangi, C. 2012. “Women, Refugees and Sexual Violence in Kakuma Camp, Kenya: Invisible Rights, Justice, Protracted Protection and Human Insecurity.” Research paper, International Institute of Social Studies. The Hague, Netherlands. Ohta, I. 2005a. “Multiple Socio-economic Relationships Improvised Between the Turkana and Refugees in Kakuma Area, North Western Kenya.” Displacement Risks in Africa, 315–37. 168 Appendixes ———. 2005b. “Coexisting with Cultural ‘Others’: Social Relationships between the Turkana and the Ref- ugees at Kakuma, Northwest Kenya.” Pastoralists and Their Neighbors in Asia and Africa 69, 227–39. ———. 2007. Marriage and Bridewealth Negotiations among the Turkana in Northwestern Kenya. Koyoto, Japan: Center for African Area Studies, Kyoto University. http://jambo.africa.kyoto-u.ac.jp/kiroku/ asm_suppl/abstracts/pdf/ASM_s37/Part1.pdf. Oka, R. 2011. “Unlikely Cities In The Desert: The Informal Economy As Causal Agent For Permanent ‘Urban’ Sustainability In Kakuma Refugee Camp, Kenya.” Urban Anthropology and Studies of Cultural Systems and World Economic Development, 223–62. ———. 2014. “Coping with the Refugee Wait: The Role of Consumption, Normalcy, and Dignity in Refu- gee Lives at Kakuma Refugee Camp, Kenya. American Anthropologist 116(1): 23–37. Omolo, N. A. 2010. “Gender and Climate Change-Induced Conflict in Pastoral Communities: Case Study of Turkana in Northwestern Kenya. African Journal on Conflict Resolution 10(2): 81–102. Opiyo, Francis Edward Omondi. 2014. “Climate Variability and Change on Vulnerability and Adaptation among Turkana Pastoralists in North-Western Kenya.” PhD thesis, University of Nairobi. Opiyo, Francis E. O., Oliver V. Wasonga, and Moses M. Nyangito. 2014. “Measuring Household Vul- nerability to Climate-Induced Stresses in Pastoral Rangelands of Kenya: Implications for Resilience Programming.” Pastoralism Journal 4: 10. Stark, L., and A. Ager. 2011. “A Systematic Review of Prevalence Studies of Gender-Based Violence in Complex Emergencies.” Trauma Violence Abuse 12: 127. Vemuru, V., R. Oka, R., Gengo, and L. Gettler. 2016. “Refugee Impacts on Turkana Hosts—A Social Impact Analysis for Kakuma, Turkana County, Kenya,” World Bank, Washington, DC. Wawire, V. 2003. “Gender and the Social and Economic Impact of Drought on the Residents of Turkana District in Kenya.” Organisation for Social Science Research in Eastern and Southern Africa. Gender Issues Research Report Series 21: 21–29. WFP (World Food Program) and UNHCR (United Nations High Commissioner for Refugees). 2014. “UNHCR/WFP Joint Assessment Mission: Dadaab and Kakuma Refugee Camps” (October). WFP and UNHCR. World Bank. 2011. World Development Report 2011: Conflict, Security and Development. Washington, DC: The World Bank. Background Paper: Implications of a G Changing Social Order and Structures for Peace and Development Context and Rationale As a semiarid, desolate environment with poor development outcomes, minimal infra- structure, and an inadequate state presence, Turkana County has long been one of the most marginalized areas of Kenya. It is significantly affected by protracted conflict: from 1997 to 2014, the northwestern subcounties registered the highest rate of fatali- ties due to violence in the country.19 In 1992, Turkana County began to host refugees, most of whom were fleeing the war in Sudan. The Kakuma Refugee Camp in northwest Turkana County subsequently expanded to accommodate people fleeing conflicts and disasters from 20 different countries in Central and East Africa. Recent crises in Soma- lia (2010–11) and South Sudan (2013–ongoing) doubled the number of refugees living in the camp, with the current estimated population to be over 160,000. Interactions between the refugees and host community have been positive and negative, resulting in incidents of conflict amid a general, overall coexistence. This background paper explores the dynamics between the host community in Tur- kana County and refugees at the Kakuma Refugee Camp; drivers of conflict that can potentially threaten peace and development; and the possibility of emerging socioeco- nomic changes, such as devolution and natural resource extraction, to exacerbate such drivers. The historical marginalization of the host community surrounding the refugee camp, communal and ethnic violence, and protracted displacement have shaped social organizations, norms, behaviors, and interactions between refugees and their hosts. Causes for concern include increasingly frequent and severe acts of violence that may escalate with the impact of devolution in counties with low institutional capacity and the discovery of oil and water, which involves an increased presence of external actors, 19. See Armed Conflict Location and Event Data Project (ACLED) data for 2014 at http://www.acleddata.com/data/. ACLED is the most comprehensive public collection of political violence data on developing states. It produces information on the specific dates and locations of political violence, the type of events, the groups involved, fatalities, and changes in territorial control. 169 170 Appendixes such as multinational corporations, nongovernmental organizations (NGOs). These anticipated development changes have the potential to either aggravate or mitigate existing conflicts. They represent opportunities and challenges to social structures and norms among host and refugee communities. Sustainable development in the region will require effective resilience and peace-building mechanisms. Methodology This paper builds on an extensive literature review followed by an analysis of qualita- tive and quantitative data on demographics, psychosocial concerns, and the perceived impact of refugees and hosts on one another. A brief exercise on data disaggregated by geographic proximity to the Kakuma Refugee Camp was conducted to determine if perceptions and concerns varied based on the respondent’s distance from the camp. Qualitative and key informant interviews were carried out with local host community and refugees that provided insights into social structures, community organizations and institutions, and conflict and violence. This paper examines existing social struc- tures and their vulnerability and resilience in relation to drivers of conflict and violence, oil and water extraction, and devolution. Additionally, it describes the mechanisms and opportunities for coping and the resilience needed to maintain peace as development initiatives continue to be carried out in Turkana County. The Turkana Host Community Turkana County is one of the cradles of humanity, with consistent and recent discover- ies showing that the Turkana region has been a center of biological and cultural evolu- tion for at least 3.5 million years. Part of the Nilotic pastoral population moving up and down the Rift Valley for the past 6,000 years, the Taker people, inclusive of the Turkana, have settled in Turkana County for over 1,000 years. Enduring the tests of time and hardships, the Turkana people have developed a rich system of traditions, beliefs, prac- tices, economies, and sociopolitical structures that define their presence in the region and frame their interactions with competing local groups.20 With 855,399 inhabitants, Turkana County is the second largest county in Kenya.21 It is located in the Rift Valley Province, bordered by South Sudan in the north, Uganda in the west, and Ethiopia in the northeast. It is among the driest counties in Kenya, receiving an estimated average of 250 millimeters (mm) of annual precipitation. Fur- thermore, rainfall patterns have changed, as long rainy seasons have become shorter 20. See “National Geographic Explorers Bios.” at http://www.nationalgeographic.com/explorers/bios/leakeys/. 21. See the Republic of Kenya’s Agriculture Sector Development Support Programme website at http://www.asdsp.co.ke/ index.php/turkana-county. G. Background Paper: Implications of a Changing Social Order and 171 Structures for Peace and Development and dryer, and the short rainy season has become longer and wetter. Overall annual rainfall remains at low levels (HRC 2015). In this way, the desolate, semiarid climate and unpredictable environment significantly contribute to the instability of the region and the lifestyles of its population, most of whom predominantly practice nomadic pas- toralism. The Turkana’s movement is limited to their subareas. The Turkana of Kakuma follow a path between the plains and the hills, travelling about 200 kilometer (km) per year within a 70–80 km radius. Only when climate disaster strikes or others push them out do the Turkana and other pastoral groups make long sustained movements away from the area (McCabe et al. 1999). The Turkana, a large ethno-linguistic group, reside in housing settlements that season- ally disperse and aggregate. These settlements have recently become more sedentary, resulting in more frequent interactions and engagement with the local community, civil society organizations, and NGOs. Many organizations aim to improve health and education in this setting, where development indicators are the poorest in Kenya. For example, only 15 percent of Turkana residents have a primary level of education, only 2 percent use electricity as their main source of lighting, only 39 percent use improved sources of water, and no Turkana residents have homes with concrete roofs.22 NGO intervention efforts aimed at alleviating years of systemic government neglect have been limited. The Turkana remain one of the most marginalized communities in Kenya, setting the stage for their interactions with other native and foreign populations such as the Pokot, the Samburu, the Karamojong, and refugees (Mburu 2002). Social Groups and Institutions There are key distinctions between social groups in Turkana society, and the types of interactions and forms of conflict associated with men, women, and youth differ. Elders play important roles as initiators and mediators of conflict. As members with their own herds and families, elders are in positions of traditional power and are often interested in expanding access to resources, facilitating commercial activities, and promoting security (Krätli and Swift 1999). They also play a crucial role in tempering conflict and controlling the behavior of youth participating in violent activities, such as cattle rustling. Women engage in important social functions as they provide for families, produce milk from livestock, care for children, and sometimes engage in trade. Although it is rare for women to use firearms, it can occur when men are away from home and there is a need to drive back raiders. In addition, women sing war songs prior to raids, which is 22. See Kenya National Bureau of Statistics at http://knbs.or.ke/index.php?option=com_phocadownload&view=category&i d=20&Itemid=1107. 172 Appendixes an important motivator in encouraging youth and men to fight for the sake of building and maintaining reputations. Women also have value in informal contexts, as some do not fully sever ties to their families after marriage. These connections are useful if women are abducted or marry into rival groups, serving as communication channels (Krätli and Swift 1999). Turkana youth are a dynamic and highly vulnerable social group. Once initiated into adulthood around age 18, youth are able to engage in activities such as war, marriage, and the accumulation of livestock and wealth. In addition, while being an “elder” was a social construction that represented an aggregate of age, political power, and economic success, these factors now take on independent significance. For example, young men who have great wealth or political authority acquired through traditional or modern means can exert similar authority as “elders” (Krätli and Swift 1999). Furthermore, core principles among youth change as opportunities for income-generation diversify. With this in mind, local development projects can generate modern forms of employment and cash income that lead to behavioral changes in youth, creating a space for conflict among peers as well as between older and younger generations. Other prevalent groups in the region include: (1) civil society organizations and NGOs with the potential to drive and mediate conflict through their delivery of aid, especially food and basic social services; (2) politicians (Members of Parliament and councilors) and administrators (county commissioners, deputy county commissioners, and chiefs) through direct involvement or by abstaining from intervening; (3) non-state security forces, often hired by the extractive industry to ensure the success of initiatives, which have become notorious for abusing and harassing local people; and (4) arms dealers and cattle sellers who profit from activities that produce and are fueled by conflict (Krätli and Swift 1999). Traditional Conflict Some conflict has always been a part of the Turkana pastoral lifestyle, but the motivations for engaging in conflict have evolved from the traditional ones. Among historical sub- sistence pastoralist populations, conflict is primarily manifested in the form of livestock raiding, a tradition of “light violence” (Unruh 2005). Often inaccurately believed to be the result of resource deprivation (Hendrickson et al. 1996), the Turkana travel along migra- tion routes that frequently move them across district boundaries and sometimes across country borders in search of water and feed for their herds. In these instances, while they compete with other groups for resources in fairly harsh environments, they also develop friendships and alliances. These relationships and social networks serve to rebuild herd G. Background Paper: Implications of a Changing Social Order and 173 Structures for Peace and Development strength and are key to famine recovery. Raiding is also associated with marriage agree- ments, rites of passage, wealth, and individual and community social status. As multidimensional and social events, raids begin with preparatory expeditions that gather intelligence on the location of livestock and the ammunition and human resources needed to execute the mission. Elders give the blessing to ensure success, while women carry out the appropriate rituals (Bollig 2000). Because livestock can be used as currency for social transactions, raiding is one of the few traditional ways to increase wealth, especially for youth who seek to garner prestige and earn income independent of their family. Livestock is also frequently used as a dowry for marriage and is reportedly a strong motivator among young men (Hendrickson et al. 1998). In semiarid and desolate areas where economic opportunities are minimal and resulting assets derived from such opportunities are constantly at risk, social and symbolic capi- tal is vital to survival (Hendrickson et al. 1996). For a population with limited access to employment opportunities and basic services, raiding represents an important chance to rebuild herds, especially following an environmental crisis. Politically, raiding serves as a medium for maintaining the separate identities of groups that engage in herding within close proximity to one another (Bollig 2000). New Forms of Violence and Drivers of Conflict Following the colonial period, the livestock economy was weakened by pacification of the region and by increased physical insecurity linked to geopolitical conflict. Preda- tory raiding, a form of raiding that starkly contrasts with raiding based on balance and reciprocity, emerged. Based on criminal logic, predatory raiding by outsiders in the form of cattle rustling has increased in frequency, largely driven by armed militias or bandit groups in Kenya and surrounding states. The primary goal of these groups is to acquire mass numbers of cattle to sell in the commercial market (Hendrickson et al. 1998). In addition, weaponry associated with raiding has increased in sophistication, shifting from spears and machetes to traditional and modern firearms accessed as a result of nearby conflicts and the regional arms trade (McCabe 1990). Several concurrent factors drive conflict and violence in Turkana County. First, minimal or absent state presence and general marginalization of the Turkana people significantly contribute to increasing conflict and violence. Without the government of Kenya’s sup- port and administration of core functions at the local level, Turkana County would lack the capacity to implement initiatives that could be accessed or taken advantage of by local residents. Security responses, criminal justice institutions involving police and judicial processes, and job growth with an emphasis on youth employment are key 174 Appendixes areas of neglect, which when addressed could be highly beneficial to local residents. Land disputes, land grabbing, and communal claims over rangeland also remain major underlying causes of conflict. Ethnic clashes with regional groups within Kenya, such as the Pokot and Samburu, and from other countries, like the Toposa of South Sudan, the Dassenech of Ethio- pia, and the Karamojong of Uganda, have increased. The increase is associated with interim periods or transition periods between scarcity and abundance (Hendrickson et al. 1998) and with pastoralist groups being pushed further toward the borders of their countries, resulting in increasing interactions in regions already burdened with high levels of conflict. The motives for raiding have also diversified, which is both a cause and consequence of exclusion from the market economy due to poor infrastructure and mobility as well as the ongoing arms trade in East Africa and the Horn of Africa. Refugees and Protracted Displacement Kakuma Refugee Camp has experienced a frequent and large influx of refugees as a result of its geographical proximity to neighboring states undergoing intensive periods of conflict and instability. Currently, according to UNHCR, refugees from South Sudan and Somalia represent the majority of refugees in the camp. The increasing population and ethnic diversity in the camp are a reflection of the government of Kenya’s policies to concentrate refugees at two main sites: Kakuma (Northwest) and Daadab (East). Kakuma Refugee Camp comprises four settlements (Camps 1, 2, 3, and 4) and is man- aged by the United Nations High Commissioner for Refugees (UNHCR) under the jurisdiction of the Department of Refugee Affairs. The government of Kenya is expected to gradually assume its full management. There are also ongoing discussions for a new camp site in Kalobeyei. Since the establishment of Kakuma Refugee Camp in 1992, the refugee population has surpassed the number of local inhabitants exponentially. UNHCR estimates that, as of September 2016, the camp is host to over 160,000 refugees, whereas as of 2009, the town of Kakuma had approximately 36,875 inhabitants.23 Some researchers have pointed out that, over the years, the Kakuma Refugee Camp has developed into a town within a town, exhibiting key features of urban settings such as high population density, basic infrastructure, and nonagricultural activities as the primary means for income genera- tion (Ohta 2004; Oka 2011; Agier 2002). 23. See UNHCR data sheets for Kenya Kakuma Camp Population at https://data.unhcr.org/horn-of-africa/documents.php?p age=1&view=grid&Country%5B%5D=110. G. Background Paper: Implications of a Changing Social Order and 175 Structures for Peace and Development Drivers of Conflict The ethnic composition of Kakuma Refugee Camp is an important driver of conflict. Ethnographic studies (Aukot 2003; Newhouse 2015) have documented how tribal ten- sions shape the political ethos of the camp. For instance, the South Sudanese Dinka- Bor and the Dinka-Bahr-el-gazal have shifted away from their traditional antagonism against each other to partner against the Sudanese Nuer in order to gain control over the Sudanese community. Tensions often triggered by low-level events have escalated into incidents of violence. Similarly, the historical marginalization of ethnic groups within their own nations, for example the Somali-Bantu or Barawa, extends into camp politics by generating continued discrimination and marginalization within the camp. This microdynamic of tribal divisions also permeates governance within the camp because members from influential tribes and clans are actively engaged in informal leadership structures that mediate conflict. Testimonies collected for this study high- light that leadership positions are rumored to be lucrative because leaders can extract regular payments from business owners in their blocks or areas. Another risk factor for violence in the camp is the growing number of unaccompanied minors and youth ages 14–24 who are unemployed or outside the labor market and not enrolled in school or vocational training. Despite existing activities targeting youth and the availability of low-cost education, many refugee youth in Kakuma drop out of school. This group lacks representation in the larger refugee community. Refugee leaders, who usually assume a role in conflict resolution, have expressed concern about reaching out to these marginalized youth, who tend to ignore advice from elders (Som- mers 2002). Finally, the psychological and emotional well-being of refugees is critical for mitigating or exacerbating the daily frustrations of life in Kakuma Refugee Camp. The region’s pro- tracted insecurity, political instability, and frequent occurrence of drought with resulting persistent famines frustrate the local population, as the time spent in limbo frustrates the refugees. There is a high prevalence of individuals spending most of their lives con- fined to an isolated camp. The situation is even more precarious because vulnerable populations, such as children and young men, constitute a large share of the refugees (Økland 2014). Symptoms of posttraumatic stress disorder (PTSD) are increasingly observed among refugees, especially among new arrivals from South Sudan fleeing the ongoing conflict there, indicating that poor mental health, if not already a pervasive issue, may increasingly become one. Some studies reveal epidemic levels of intimate partner violence and other forms of violence against women and girls as contributors to poor mental health (Horn 2010). 176 Appendixes Refugee–Host Dynamics and Their Impacts on Social Structures and Organizations Kakuma Refugee Camp and its surroundings exist in a harsh and arid environment coupled with poor infrastructure and limited access to basic services. As nomadic pas- toralists, the Turkana host community depends mainly on livestock for survival. With their lifestyle weakened by livestock disease, famine, and ethnic clashes with neigh- bors—mainly the Pokot and Karamojong—a large presence of refugees in a marginal- ized area with a limited presence of state institutions is a critical challenge for both the host and refugee communities. The presence of a refugee camp affects the host population positively and negatively (Jacobsen 2005; Crisp 2000). The impact of refugees varies according to gender, class, and the host community’s regional and generational characteristics. Therefore, context- specific analysis is needed to determine who benefits, who loses, and why (Whitaker 1999; Las sailly-Jacob 1993). An important feature of the Kakuma context is that the refugees in Kakuma Refugee Camp do not have any type of sociocultural affinities with the host community in Turkana, which differs considerably from many camps of recent decades located in bordering areas in Africa where refugees had close ethnic and social relations with their hosts (Horst 2001). Furthermore, the marginalized and isolated condition of Turkana communities circumscribed their sociocultural interactions and development opportunities to within their own country. The Turkana’s attitude toward the refugee camp has been ambiguous and contradictory (Ohta 2004). Since the initial phase of the camp’s establishment, the host community was apparently aware of the benefits, mostly economic, from the presence of refugees and international humanitarian assistance (Oka 2014; Jansen 2008). However, there have also been complaints about the insecurity and environmental damage caused by the camp. For example, a restriction on collecting environmental resources has not been established, which has enabled violent confrontations between hosts and refugees. As a marginalized group in Kenya, the Turkana have actively worked to maintain control over environmental resources in the area around the camp, partly through negotiations and partly through the use of violence. The economic interactions between displaced people and their Turkana hosts have multiplied and range from employers and employees, partners in petty trade, and bond- friendships arising from the exchange of gifts or marital-like relations. These practices have evolved over the years, with local elders playing a key role in using customary law to mediate disputes between the Turkana and refugees. In this regard, UNHCR and the Department of Refugees Affairs have been instrumental in guiding and advising G. Background Paper: Implications of a Changing Social Order and 177 Structures for Peace and Development refugees on conflict resolution through traditional leaders. Economic activities have greatly influenced the interaction between and even interdependence of refugee and host. Given their pastoralist tradition, many Turkana have entered into contracts with refugees regarding livestock. In some cases, locals have herded cattle owned by refu- gees, with refugees acting as brokers between locals and butchers at the camp (Ohta 2005). Certainly, the economic dynamics of the Kakuma Refugee Camp have impacted local institutions and relationships as many Turkana women, young boys, and girls are involved in informal and petty-trade activities, ranging from housekeeping, to collect- ing firewood and charcoal, to selling milk in the camp. This dynamic has changed the nature of income sources for host households as women and children have become important providers for their families. Nevertheless, the informal nature of the econ- omy within and around the camp also negatively affects the conditions of a new and unprotected labor force. Similarly, communities have become dependent on the exchange and resale of the aid rations received by refugees (Keen 1994; Newhouse 2015). Several researchers have documented how the Kakuma ration resale market operating near food distribution centers not only shapes the everyday lives of refugees and survival mechanisms, but also how they coexist with host communities (Oka 2014). Another form of socioeco- nomic interaction widely examined in ethnographic studies is the bond-friendship—a customary practice among the Turkana in which two individuals, together with family members on both sides, establish a close relationship through the exchange of gifts. The relationship begins with the exchange of small gifts, such as tobacco, and continues with each party regularly visiting the other’s homestead and the giving of a goat or a sheep, which is a typical way of extending a cordial reception. Eventually, an interdependent relationship is built, which is very important for the Turkana (Ohta 2005). Bond-friend- ships have proliferated between refugees and hosts, creating new ties based on custom- ary practices. At the same time, economic interactions have influenced local culture, especially for young Turkana women involved in domestic work and petty trade. Some have become multilingual, including learning KiSwahili,—a lingua franca—in the camp. Social relations have been strengthened through intermarriage between the Turkana and refugees, usually involving a Turkana woman marrying a refugee. While this is a positive aspect of the complex interactions at Kakuma Refugee Camp, some con- flicts have arisen related to the emergence of “spontaneous marriages,” or marriages not grounded in traditional practices in which, following extensive discussion, two extended families reach an agreement over the dowry of the bride. For some refugees, there is a reluctance or limited economic capacity to follow traditional dowry rules. To a certain extent, conflicts are caused by the refugees’ lack of knowledge and acceptance of local norms as well as their perceptions of living in a temporary status despite the 178 Appendixes fact that most are living in protracted conditions. Therefore, new mechanisms have emerged to mediate conflicts associated with marriage relations, usually through local elders and camp officials. Other studies have also mentioned that sexual exploitation of women and girls has increased due to the presence of refugees and foreigners, with serious consequences for the social fabric and cultural bonding of the communities (Mwangi 2012). New Dynamics between Refugees and Hosts Quantitative data was collected from the host population, sampled across four loca- tions: the towns of Kakuma, Lokichoggio, Lorengo, and Lorugum. Kakuma town is closest in proximity to Kakuma Refugee Camp (4.4 km); Lorugum is the farthest (171 km); and Lorengo and Lokichoggio are in the middle at 25.5 km and 97.6 km, respec- tively. Disaggregated by proximity to the camp, the data reveal that there are key differ- ences and similarities among sampled sites with regard to psychosocial concerns and perceptions of refugee impact. Many of the positive impacts were observed in relation to economic opportunities and service delivery (figure G.1.). However, supplemental qualitative data show that eco- nomic interactions are complex due to their informality and their tendency to override fundamental Turkana values of balance, reciprocity, and trust. Participants surveyed in the town of Lorengo, followed by those in Kakuma town, most frequently reported that the presence of refugees were good for trade. It was also noted, most frequently by Lorengo and Kakuma town residents, that refugees cheat the local Turkana in trade, which highlights the complex nature and nuance of everyday economic interactions between host and refugee communities. Figure G.1. Perceptions Regarding the Economic Impacts of Refugees 45 Kakuma Town 40 Lokichoggio 35 Lorengo Percent of Turkana 30 Lorugum 25 20 15 10 5 0 Good for trade Refugees bring Refugees cheat Refugees Refugees get NGOs don’t care resources/services Turkana in trade rob/steal resourcesTurkana about Turkana from Turkana should get Source: Vemuru et al. 2016. G. Background Paper: Implications of a Changing Social Order and 179 Structures for Peace and Development Congruent with the literature, the most common economic interactions are related to the trade of food aid, youth and female employment in domestic work, and cattle herding. These interactions are not necessarily negotiated on the basis of reciprocity and trust (figure G.1). Although the economic dynamics are presumably based on interdependence, the Turkana tend to be in a more vulnerable and dependent situ- ation because their economic opportunities from other sources, including extractive industries, is marginal. The refugee economy represents their main source of income and service delivery. Many Turkana, primarily those in Lorengo, feel that refugees rob and indirectly steal from them through the taking of land and water, by using resources from the Kenyan government and UNHCR, and by underpricing labor and production. The belief that refugees receive resources that local Turkana deserve was stronger in Lorugum than in any other location despite the fact that it is the farthest distance away from Kakuma Refugee Camp and therefore only minimal daily interaction of its residents with refu- gees can be assumed. The belief that refugees receive an undeserved amount of services is most likely based on perceptions rather than reality because for over 20 years, Loru- gum has been the recipient of sustained agricultural and pastoral development associ- ated with the Turkwel River Power Project (Oka 2014). The negative impacts perceived by hosts were more related to sociocultural issues associated with the presence of refugees (figure G.2). Overall, hosts perceive refugees as good neighbors, but this seemingly positive relationship includes several contra- dictions. For example, hosts frequently reported a sense that refugees kill and harass local Turkana. Other studies show that refugees feel the same about their hosts. In fact, Figure G.2. Perceptions of Refugee Social Impact Kakuma Town 30 Lokichoggio Lorengo 25 Lorugum Percent of Turkana 20 15 10 5 0 Refugees Increase/ Prostitution Refugees Refugees Refugees Conflict as friends/ introduction harass sexually kill (general) neighbors of drugs/ Turkana exploit Turkana alcohol Turkana Source: Vemuru et al. 2016. 180 Appendixes there have been many incidents of community violence around the camp over the last decade, with small, low-level tensions rapidly escalating into large community and eth- nic clashes. For example, a problem related to a cattle transaction between a Turkana and a refugee can easily develop into a violent fight. Given the ethnic composition of the camp, conflict can be even more complex and build on preexisting rivalries among specific groups, such as the Sudanese Dinka and the Somali-Bantu. Additional key social impacts are related to sexual exploitation and informal, marriage- like relations, which do not align with traditional marriage institutions and associated economic relations, such as dowries. These issues affect gender roles because the more informal relations undermine the economic function of marriage. In this regard, refu- gee–Turkana “marriage” relations are weak, due not only to different sociocultural val- ues, but also to ideas of transition and settlement status among refugees. Some refugees do not willingly assimilate because of their denial regarding the protracted nature of their displacement or because they are forbidden to settle down for fear of losing their status as transitory refugees. As many hosts have noted in focus group interviews, “they [refugees] are not Turkana, and they didn’t come to stay.” Surprisingly, high levels of substance abuse is reported. The increased use of drug and alcohol use is an important risk factor for the mental health and psychosocial well-being of refugees and hosts. This issue should be examined in further detail, because it is also a risk factor for multiple forms of violence, including intimate partner violence. With new opportunities for employment and economic growth emerging, potential beneficiaries may find it difficult to access or persist in meaningful opportunities if they have poor psychosocial health as a direct or indirect result of experiencing drug and alcohol abuse. A psychosocial stress analysis in conjunction with nutritional indicators demonstrate that the number of listed stress factors are correlated with nutritional status. Hence, the almost identically high rates of nutritional status in Kakuma and Lorugum correlate with high and low stress indicators, while the similarly low rates of nutritional status in Lokichoggio and Lorengo demonstrate a middle range of psychosocial stress factors with a smaller standard deviation (figure G.3). We interpret this finding to suggest that while the presence of refugees at Kakuma adds to the stress of the host community, it also enables improved nutritional access through labor and exchange and through development. Lorugum, which has had sustained development for over 20 years, has a high rate of nutritional status, but with a lower rate of reported stresses and concerns. Interestingly, the overlap of concerns between Kakuma and Lorugum seems to be related to issues such as education, the economy, and employment opportunities. G. Background Paper: Implications of a Changing Social Order and 181 Structures for Peace and Development Figure G.3. Concerns of Hosts Across Sampling Locations Kakuma Town 100 Lokichoggio 90 Lorengo 80 Lorugum Percent of Turkana 70 60 50 40 30 20 10 0 t irst r ym t/ life ing ss s ees g s ep ntal ugh nge ue tie idin on t plo en e en erd Illn Th ol f iss cili rd leti rem ploym n/d nme Hu Dro Ra ha l fa orh ee ho ty/ da nviro fug f sc ica df un em ver ed Re Lan ko tio Un E Po fm de Lac ko gra Lac de Source: Vemuru et al. 2016. The psychosocial stress felt by the Turkana in Lokichoggio and Lorengo is corre- lated with a middle range of concerns, which includes hunger, drought, and water for animals. Feelings of resentment toward the government resulting from a sense of being continually disadvantaged and marginalized (matharau) is usually the primary concern of host community members. Economically, their key concern is related to hunger and drought (figure G.2), indicative of the limited resources available for their survival—directly in the form of food supply and indirectly in the form of livestock maintenance. Host communities have historically used conflict tools such as livestock raids to renew their supply of livestock after loss due to famine, drought, or raiding. With increasingly available information on climate change, environmental stressors may drive conflict even further into Turkana as resources such as grazing land and water become scarce. Although refugees are not involved in this predicament, limitations on supply derived from pastoralist lifestyles may exacerbate or contribute to low-level tensions associated with the presence of refugees and could escalate into serious incidents of conflict and violence. Other expressed concerns include thirst, underemployment, illness, and lack of medi- cal facilities. Thirst reflects either a low availability or limited access to a water supply, including bore wells in and around Kakuma Refugee Camp, which qualitative data sug- gest is associated with instances of conflict. Concerns about thirst are higher in Kakuma 182 Appendixes Figure G.4. Plot of Principal Components 1 and 2 Note: Line lengths represent the frequency with which a concern was reported. Angles between lines correspond to correlation coefficients. L1=Kakuma town; L2=Lokichoggio; L3=Lorengo; L4=Lorugum. Source: Vemuru et al. 2016. town than in any other sampling location. Underemployment was high in Kakuma town and Lorugum and present in Lokichoggio and Lorengo. Illness is a key issue in Kakuma town, Lokichoggio, and Lorugum, but not in Lorengo. Data from Lorengo on the perception of refugees as being good for trade may signify that residents of Lorengo are better equipped economically and thus experience illness at insignificant levels. Illness was more frequently reported as a concern in Kakuma town, but corresponding reports of concerns regarding a lack of medical facilities were few; the opposite was true for Lorengo. A detailed principal component analysis allows us to disaggregate concerns that vary in frequency between locations from ones that were either very rarely reported or ubiqui- tously reported. In this way, the analysis offers a quantitative and visual representation of the concerns experienced by residents from each location, allowing us to focus only on those in the sample that vary significantly. Figure G.4, which plots PC01 against PC02, shows a cluster of concerns that are rarely or sporadically reported and several others that are more frequently reported. Although it may appear that hunger (concern code 30) is an important concern to consider due G. Background Paper: Implications of a Changing Social Order and 183 Structures for Peace and Development Figure G.5. Plot of Principle Components 2 and 3 Note: Line vectors represent the degree of association with location. Angles between lines correspond to correlation coefficients. L1=Kakuma town; L2=Lokichoggio; L3=Lorengo; L4=Lorugum. Source: Vemuru et al. 2016. to its frequency, it is in fact reported so ubiquitously in each location that it does not tell us anything meaningful about how the locations vary. Based on this P, the follow- ing concerns are salient factors to consider (concern codes in parentheses—details in appendix C): drought (10), general illness (40), illness of children/family (42), illness of livestock (43), lack of medical facilities (44), unemployment/underemployment (50), poverty/difficult life (51), lack of school fees (52), and raiding (60). Figure G.5, which plots PC02 against PC03, draws out important differences between locations. In this plot, it is important to note that line vectors represent the degree of association between particular concerns and the various locations in the sample. Based on the covariances that the principal component analysis reveals, it is possible to narrow down the concerns further by lumping together concerns that closely co-vary in both PC01/PC02 and PC02/PC03 plots. Two clusters of concerns fit these criteria: (1) illness of children/family, illness of livestock, and lack of medical facilities; and (2) unemployment/underemployment and poverty/hard life. We can refer to the first cluster as concerns about other-oriented health as opposed to the generalized illness concern that does not co-vary with this cluster. The second cluster can be classified as economic hardship concerns. We now have a list of seven concerns that are statistically meaningful for the analysis: 184 Appendixes 1. Drought (10) 2. Thirst (11) 3. General Illness (40) 4. Other-oriented health (42, 43, 44) 5. Economic hardship (50, 51) 6. Lack of school fees (52) 7. Raiding (60) Importantly, figure G.5 reveals several key differences between Kakuma and Lorugum that do not appear in figure G.4. Economic hardship concerns are most frequently reported in Kakuma, as is thirst, but only slightly more frequently than in Lorengo. Drought and other health-oriented concerns appear to be most salient in rural Lorengo. Lack of school fees and general illness are the primary concerns reported in Lorugum. However, illness is also a significant concern in Lokichoggio, and its close proximity to the South Sudanese border makes raiding the greatest source of concern at this location. Understanding the Narrative of Distrust, Violence, and Uneasy Coexistence Beyond quantitative data, qualitative analyses and discussions with the Turkana and refugees on perceptions of the other with regard to violence and conflict evokes many salient themes. In individual and small group interviews, refugees, primarily from Ethiopia and Somalia, report being fearful of the Turkana. Two refugees interviewed for this study report the following incidents: “Turkana men stole bags, a bicycle, and food. Police arrived on the scene after one and a half hours and provided minimal support. Unguarded gates and minimal protection have led to organized patrol groups to carry pangas [machetes] and stones as defense. Demonstrations by refugees related to the lack of protection have further escalated into tensions with the police. I live in fear every night that someone will attack, because women are being raped. Sometimes people are not injured if no one resists. There is no police response, and no arrests are made, going so far as to deny evidence of Turkana identity. People are supposed to keep quiet and have no protection, except to shout to neighbors, throw stones, and try to keep lights on. No one knows and understands their suffering.” (KII 2015) In focus groups, refugees reveal that violence is mostly a problem at night. The minimal resources available to use for protection and self-defense is a strong concern among attacked groups. Some refugees report feeling sympathetic toward the Turkana, while G. Background Paper: Implications of a Changing Social Order and 185 Structures for Peace and Development others view the Turkana as uncivilized. One refugee, reflecting the views of many, illus- trates the high incidence of reported theft, both in their homes and in the marketplace. “Initially [they] were okay with host community. Killings began in 1994 when local Turkana would come into the camp at night to rob refugees. Most vio- lence was associated with robbery and looting.” (KII 2015) Some refugees express a lack of sympathy toward the Turkana despite recognizing their poverty and paucity of resources. But some refugees hold the opinion that if the rela- tionships are carefully handled, the Turkana are not “all bad,” and many are “good” for trading and supplying meat. “We normally hold them like eggs,” says one Ethiopian ref- ugee about their careful interactions with local Turkana and how they cultivate positive relationships with their host community. Lastly, the majority of refugees interviewed recognize that at least to some degree, the camp has helped the Turkana, with a few suggesting that increases in income and reduction in hunger has resulted in decreased levels of violence. Interestingly, the views of the local Turkana were often quite similar. Reflecting on how the level of conflict easily escalates, one says, “We are afraid of the refugees because they like fighting. If you hit someone by accident with your bicycle, you might be beaten by a mob.” Another participant agreed, remarking that, ‘’A small incident will cause a big fight, but the refugees can also be very good to you because they give us business and we get cheap stuff at the camp.” Another participant blamed miscommunication as a frequent cause of conflict, remarking: “When conflicts start, they start from misunderstandings on a fight or mur- der. That’s how they start. It’s always something small. And there is so much ng’imeny, matharau24 that it builds up. Even the refugees have their own sense of matharau. So everyone has lots of things to be angry about.” (KII 2015) Many Turkana hosts recognize that the problem does not lie with refugees but instead with their ability to access services, such as health and education. One refugee com- ments on the neglect from the government of Kenya by saying: “The main problem is that the host community has been suffering from con- flict and neglect by the Nairobi government, so their plight is like that of the refugees. But the host community does not enjoy the same access to health and education. This causes many problems.” (KII 2015) 24. A sense of resentment and of generally being dismissed that is experienced by many local Turkana. 186 Appendixes The issue of minimal food and water and the difference in treatment between refugees and host communities is mentioned by a Turkana man: “Mostly there was no problem, our people would complain about the lack of food and water because the refugees were given for free.” Thirst, an issue that was reported with high frequency in the quantitative data, is repeatedly cited as a clear illustration of the government’s neglect of the Turkana people: “The refugees get water from UNHCR, and use that water to grow their veg- etables so they do irrigation and agriculture in their backyard. We have faulty pipes, broken pumps, and generators that break often. When I see that the refugees are getting water and I am thirsty, it doesn’t feel good. But I don’t blame them. I feel bad because I have to pay a huge price for one jerry can of water. But many of the refugees are good, you can make friends with them, and they will help you.” (KII 2015) Others continue to emphasize that systemic neglect by the government of Kenya and inequity in receiving resources from the government, UNHCR, and donors are signifi- cant factors contributing to matharau. One local host remarked: “How can I think that this is my country when as a host, I am begging the refugees for survival? I am a guest in my own country, the country belongs to the down-Kenyans.” Access to food is another main concern. During an interview, one participant empha- sized the fact that “the Turkana people do not like commerce, but hunger speaks,” which illustrates both a reluctance and need to interact with refugees. Food represents an important advantage for families. Addressing the economic value of food versus money, one Turkana claimed: “Food is much better than money, because we can keep some to eat, I can manage food well, nine bowls of sorghum last week I still have leftovers. But 100 shillings, it goes like that, very quickly.” (KII 2015) Many Turkana participants pointed out that hunger—or just a general need for food—as well as periods of conflict or coexistence are strong drivers of interactions with refugees. One explained, “Refugees, the people we work for, treat us well. Sometimes, even when there is no work, they’ll give us food. Some are good, and some are bad. Some harass us a lot.” G. Background Paper: Implications of a Changing Social Order and 187 Structures for Peace and Development Emerging Socioeconomic Changes in Turkana With experiences of conflict and marginalization that predate the arrival of the camp, the Turkana have limited access to development resources, including humanitarian aid and assistance to refugees. However, new development opportunities are emerging in Turkana County, namely the extraction of oil and water and the devolution of national government authority, including funding to the local county. If local governments can manage these new opportunities and transparently provide adequate services, they may be able to mitigate rising conflicts. Extractive Industries As frequently illustrated in the history of modern Africa, extractive industries have a significant impact on the development of poor nations. Many countries with an abun- dance of nonrenewable wealth, such as minerals and fuel, experience minimal economic growth and poor development outcomes. This “resource curse” or “paradox of plenty,” can be the result of government mismanagement of resources or weak, ineffectual institutions. Therefore, understanding the role of the extractive industries in Turkana and the impact they have on the current state of fragility and the devolution process is essential. In Turkana, extractive industries present a challenge for host communities for multiple reasons. In many cases, Turkana have noted that while employment in oil and water extraction is desirable and advertised, it is not easily accessible or beneficial to local residents. One Turkana remarked on the management of employment opportuni- ties for Turkana, saying: “I feel sorry for the wazungu, who sit in these NGOs. Then they think that we are poor because we are not smart, because we are old-fashioned pastoralists. But we are smart enough to know when things do not work out because the wakubwa (big ones) have consumed any benefits that could come to us.” (KII 2015) With the oil industry creating tension among Turkana’s subcounties, there is mounting concern regarding tenders and jobs, including growing perceptions that employment opportunities are not benefiting locals as expected in terms of quantity and quality of jobs. One focus group participant explained: “Devolution is a good thing, but only if leaders are not corrupt. We could get jobs for the oil companies, construction, but I find it risky. What if I go there, to find a job, and they only hire people they know or people who can pay?” (FGD 2015) 188 Appendixes A constricted job market, which has limited the Turkana to unskilled labor positions and prevented them from attaining management roles in processes and programs directly affecting them, has generated frustration among local residents: “I was there in Lokichoggio at the time. The management of these compa- nies—look at the names, from the top to middle or even the clerks. You won’t find Turkana there. You’ll find them as watchmen, cleaners, or road marshals.” (KII 2015) Because the Turkana can only acquire low-level positions, many do not see an oppor- tunity to or the benefit of fully transitioning from traditional pastoralism to modern forms of employment, as the Kenyan government encourages. An estimated half of oil company employees are local to the area, and their participation is mainly in unskilled and semiskilled labor activities. This can be explained in part by lagging edu- cation indicators and limited citizen participation in negotiations related to resource extraction concessions, including jobs, social services, and infrastructure that benefits communities. Additionally, potential conflict over the extraction of nonrenewable resources may be linked to limitations of physical space for herding livestock. With the government’s offi- cial announcement of the existence of viable quantities of oil for commercial purposes in Turkana, Tullow PLC/Africa Oil in northeast Turkana and CEPSA in south Turkana, the main companies operating in the county have stopped exploration and are planning extraction. A preliminary analysis of the impact of current oil activities on communi- ties (Mkutu and Wandera 2015) reveals that investments or oil installations frequently displace pastoralists from important grazing sites, migratory routes, and water sources. This development-induced displacement has sparked community violence and reani- mated historical rivalries between pastoralist tribes in northern Kenya, particularly among the Turkana and the Pokot, which could rapidly evolve from livestock-raiding conflicts into a more complex border and land conflict. As one Turkana notes, “the inse- curity in South Turkana right now is not because of cattle, it’s because of oil,” illustrating that development-induced displacement may be a larger threat than the traditional, routine “light-violence” of livestock raiding. Similarly, water discoveries not only suggest a wealth of new opportunities for local people, but also many challenges requiring sensitive management of the exploitation of national resources in order to fulfill the expectations of locals regarding benefit-sharing. One local Turkana resident commenting on the importance of water and its vital role for employment and well-being, says: G. Background Paper: Implications of a Changing Social Order and 189 Structures for Peace and Development “We have good and fertile soils here, the problem now is water. If we can get enough water, we will be able to do a great job with agriculture. But I remember that even though our field was a small field, it was enough to feed the family, to welcome the in-laws, and even buy animals.” (KII 2015) RTI/UNESCO has reported that while the Turkana aquifer water resources could serve as a water reserve for the national population of 41 million for 70 years, fears are wide- spread that the newly discovered resources are destined to be removed from the area, as happened with the Turkwel Dam’s hydropower. These fears are manifest as demonstra- tions and local unrest, fueled by narratives that the power plant has diverted power to other parts of Kenya and away from local beneficiaries. The historic marginalization of Turkana County and its people and their mounting frustration over the lack of response from local and national authorities with respect to extraction-related revenue allocation are constant drivers of conflict, and will remain so until effective local management is in place. Local residents note that misinformation is a big problem: “After water, the leaders are the biggest problem. They are the largest con- straint on us, even if we get water. Even right now, there is a county budget with money for everything, but nothing is happening. There is a lack of infor- mation, people don’t know their rights. That is the actual reality.” (FGD 2015) Building the capacity of institutions in the context of the devolution process should also focus on addressing potential grievances from distribution of revenue from resources; which could mitigate against these perceptions. Devolution In addition to a dynamic interplay between host and refugee and important challenges associated with emerging economic opportunities, Kenya’s devolution is one of the most ambitious institutional reforms in the world, involving political, fiscal, and administra- tive decentralization (World Bank 2014). The 2010 Constitution of Kenya created 47 counties, each with an elected governor and a county assembly with public finance and service-delivery mandates. The constitution’s strong emphasis on citizen participation and transparency is very important for social accountability. The legal framework for devolution includes multiple provisions requiring national and county governments to engage citizens in policy making, planning, oversight of public resources, and acces- sibility of information to the public. The constitution further stipulates a greater role for parliament and an independent judiciary. 190 Appendixes Kenyan society has embraced devolution as a vehicle for achieving the country’s devel- opment objectives, including the delivery of health, urban, and agricultural services. In interviews with local residents, one Turkana notes the following: “I really want the [government] to help us with agriculture. If I can grow enough to feed my children, and also to buy clothes, uniforms, pay fees, then I don’t have to go to the camp anymore, don’t have to suffer from ng’imeny when I implore the refugees to buy my charcoal because my children are starving at home.” (KII 2015) While governors have become central actors working to fulfill this promise and have carried out multiple initiatives to deliver tangible results, they face serious institutional challenges. A lack of coordination and clear division of functions among county and national governments is affecting the devolution process, as is the uneven distribu- tion of human capacity and financial resources across counties. Additionally, several social groups, particularly youth, perceive that devolution as it has been implemented so far will only work in counties with effective management, and other counties will lag behind (World Bank 2015). Some local Turkana residents mention a lack of account- ability and county neglect as key obstacles to successful development programming: “Well, devolution does not mean that we get better leaders. Yes, the county will have more autonomy, but now with less accountability from the center.” “The refugees are far more helpful than our government. When they give us food, we wonder why it is that our own government neglects us.” (KII 2015) Other express concerns include a lack of participatory mechanisms to ensure engage- ment—not only of youth but all citizens—in planning and overseeing resource alloca- tion to counties. The development of more inclusive and accountable policy-making instruments is critical to sustainable devolution. Another important aspect related to decentralization is that historically marginalized counties around arid and semiarid areas like Turkana face the greatest opportunity and at the same time the lowest institutional capacity to benefit from devolution dividends. While the largest per capita transfers correspond to marginalized counties because of their low population density and lack of resources, they also have major needs for invest- ment, including roads, transport, human development, and resource management. Technical support and tailored interventions aimed at improving institutional capacity in marginalized counties will therefore be essential in helping these lagging regions take advantage of the devolution process and reach average national development levels. G. Background Paper: Implications of a Changing Social Order and 191 Structures for Peace and Development Despite recently garnered authority and funding from the national government, local institutions are unable to effectively manage county resources, which could trigger an increase in the frequency and severity of conflicts. Conflict and violence are critical challenges to devolution in marginalized counties, several of which are in bordering areas affected by protracted conflict and displace- ment, such as the Karamojong cluster in Uganda. A 2008 Small Arms Survey in Tur- kana revealed that nearly half of all respondents had witnessed a violent event in their lifetime (McEvoy and Murray 2008). According to ACLED data, most fatalities related to violence in Kenya from 1994–2014 have been registered in the northwestern county of Turkana (1,233 out of 8,500 nationwide) with a downward trend after 2000.25 The County Police Authorities (CPAs) created by the National Police Service Act will serve as an important interface between counties, communities, and the National Police Ser- vice (Mktutu and Wandera 2014), and they are expected to improve the management of security services at the local level. However, new forms of violence, including organized crime, trafficking, and violent extremism, stress formal and traditional institutions of conflict mediation and community peace-building. Anecdotal evidence demonstrates that ethnic tensions resulting from competition for resources, cultural practices, and contested political representation are approaching crisis levels with devolution, which can partly be explained by the culture of politi- cal patronage and marginalization within counties. Devolution in Turkana County has reportedly made people more aware of the historical injustice and marginalization that their ethnic group has experienced (Mktutu and Wandera 2014). This perception influences their relations with the national government, constraining the constituency of emerging multiethnic governance structures. Because this region is host to a large refugee population, including those at Kakuma Refugee Camp, understanding the impact of devolution on refugee policies and programs is vital. The Danish Refugee Council (ReDSS and Samuel Hall 2015) explains the discourse at the county level on refugees reflects varying views: on one hand, counties want to be engaged in refugee affairs because they must deal directly with the consequences of hosting refugees. On the other hand, county authorities feel that refugees are a burden and that the national government or UNHCR should compensate the county for its responsibility and man- agement of refugee issues. Discussions regarding the role of local authorities in finding durable solutions to displacement through territorial development are incipient, and additional efforts are needed to develop planning and budgeting capacity throughout the devolution process. 25. See ACLED database at http://www.acleddata.com/data/. 192 Appendixes Local Resilience and Conflict Prevention County governments in Kenya are well placed to prevent conflict and violence (Agade et al. 2014). They already support initiatives that have demonstrated positive results in mediating conflict and interrupting the perpetuation of violent cycles and have estab- lished peace committees to address insecurity within and among counties. Similarly, peace agreements, such as the Lokiriama Peace Accord signed between the Turkana and the Karamojong, are structures that can help counties strengthen coordination with the national government. A peace ambassador was recently appointed in Turkana to coordinate stakeholders and organize mediation efforts. Conflict resolution interventions have been increasingly supported by UNHCR and other organizations. Some programs have focused on security, such as the implementa- tion of community policing to promote improved dialogue between communities and the police and the creation of mobile courts to ensure access to justice and protection of refugees. Other actors support the participation of women and youth in leadership structures as a way to prevent community violence. There are currently youth groups in the camp jointly led by male and female representatives elected under the supervision of the Department of Refugee Affairs. In addition, there are youth forums aimed at pro- moting peaceful coexistence that include the host community (UNHCR 2015). These new constituencies have the potential to lead to more inclusive and intergenerational leadership to mediate social tensions and work toward durable solutions to displace- ment through community resilience. Youth-led peace-building initiatives have recently emerged, particularly “Peace Cara- vans”—local programs led by young men and women from pastoralist communities, including the Turkana, the Pokot, and the Samburu, who recognize the neglectful state presence in northwest Kenya and who seek to resolve violent disputes through com- munity dialogue. Youth travel as a cohesive unit to areas of conflict and high tension to provide communities with a space to vent their frustrations and search for amicable solutions. Their efforts, supported and assisted by the Kenyan government, have con- tributed to the cessation of violence between historical rivals (for example, the Pokot in Baringo East and the Turkana in Turkana South). These peace-builders have even devel- oped partnerships with local elders, women’s groups, community-based organizations, and local administrative institutions to implement peace agreements. For example, a Turkana member of the Laikipia Peace Caravan helped draft the “13 Commandments” for fighting cattle rustling, which were included in a 2010 peace agreement, enabling rival communities to share water and pasture resources and reduce conflicts related to resource-scarcity. G. Background Paper: Implications of a Changing Social Order and 193 Structures for Peace and Development Conclusion Extensive literature and data support the notions of Turkana County as a territory that is extremely marginalized, fragile, and underdeveloped. Poor development outcomes, minimal infrastructure, and a neglectful state presence are critical challenges that must be addressed to prevent an increase in the severity and frequency of conflicts and violence. The protracted nature and large size of the influx of refugees are signifi- cant aspects of the Turkana landscape, and an understanding of them is vital. Since the establishment of Kakuma Refugee Camp, the refugee population has exponentially sur- passed the number of local inhabitants: UNHCR estimates that as of September 2016, the camp is host to over 160,000 refugees, whereas as of 2009, the town of Kakuma had approximately 36,875 inhabitants.26 Interactions among the host community and refugees have been positive and negative. Since the initial phase of the camp’s establishment, the host community was aware of the mostly economic benefits of the refugee presence, but many complained of insecurity and environmental damage caused by the massive and continuous influx of refugees. Data disaggregated by proximity to the camp reveal that there are key differences and similarities among sample sites in terms of perceptions of the impact of the refugee presence. Of the Turkana interviewed at four locations—Kakuma, Lorengo, Lokichog- gio, and Lorugum—41 percent claimed that when entering the camp to trade, they have experienced a combination of mistreatment, economic malfeasance, and violence at the hands of refugees. In Kakuma town, adjacent to the refugee camp, 33 percent reported refugee violence and mistreatment. As distance from the refugee camp increases, the proportion of people who believe that refugees have a positive impact on Turkana decreases. The Turkana do not believe they have been able to fully exploit the economic oppor- tunities offered by their interactions with refugees, which range from partners in petty trade to bond-friendships to employers and employees, because of their poor access to education, limiting them to unskilled labor positions. Feelings of resentment resulting from an experience of being continually disadvantaged and marginalized—marathu— are among the most frequent worries faced by the host community. One interviewed Turkana explains: “How can I think that this is my country when as a host, I am begging the refugees for survival? I am a guest in my own country, the country belongs to the down-Kenyans.” 26. See UNHCR Data sheets Kenya Kakuma Camp Population at http://data.unhcr.org/horn-of-africa/settlement.php?id=17 &country=110®ion=12. 194 Appendixes While the ambitious devolution process is a well-intentioned and promising institu- tional change to transform historical marginalization, it is still seen as a distant promise in contexts of low capacity for administering and managing recently allocated funds and authority. Furthermore, if devolution is not fully developed and if it does not strengthen the presence of the state in Turkana County in a transparent and accountable manner, the existing marginalization, ethnic divisions, local grievances, and conflicts could be exacerbated. Tensions between the national and local governments could also impact the contradic- tory and dynamic relations between refugee and host because historical marginaliza- tion constrains the host community’s access to and participation in benefits from new development opportunities. Extractive industries bring with them a potential source of conflict that could threaten the stability of the region, especially if local authorities and industry vendors lack mechanisms for the transparent, meaningful, and active partici- pation of communities in the negotiations and allocation of revenues.27 Land disputes and communal claims over rangelands that result from the presence of external com- mercial actors also pose threats to social and economic stability. Ongoing discussions regarding extending Kakuma Refugee Camp into Kalobeyei is a key issue requiring close examination. It is important to strengthen local consultation structures in refugee and host communities to ensure participation in the planning and development of the new camp, including a clear explanation of the economic and social risks and opportunities associated with the protracted nature of displacement and a potentially enlarged presence of refugees in the area. Drivers of conflict in Turkana County range from host–refugee interactions to the changing nature of nomadic pastoralism, and from climate change to interlinked forms of violence and protracted conflict in neighboring areas. Opportunities for building on dimensions of local resilience include consolidating local institutions, such as tra- ditional mediation mechanisms that maximize the role of local leaders among elders, youth, and women, and improving the presence of key justice and security institutions. A territorial development approach is crucial to ensuring steady economic growth, social inclusion, and peace consolidation in Kenya, as well as stability and peace-build- ing efforts in the region. 27. Monitoring the role of vendors will be an important part of preventing future conflict because in the past, they have been associated with harassment and abuse of local residents and what is often referred to as the “resource curse.” G. Background Paper: Implications of a Changing Social Order and 195 Structures for Peace and Development References Agade, Kennedy Mkutu, Marani, Martin, and Mutuma Ruteere. 2014. Securing the Counties: Options for Security after Devolution in Kenya. Nairobi, Kenya: Centre for Human Rights and Police Studies (CHRIPS). http://www.chrips.or.ke/docs/publications/Securing_the_Counties-_Options_for_Secu- rity_after_Devolution_in_Kenya.pdf. Agier, Michel. 2002. “Between War and City Towards an Urban Anthropology of Refugee Camps.” Ethnography 3(3): 317–41. Aukot, Ekuru. 2003. “‘It Is Better to Be a Refugee Than a Turkana in Kakuma’: Revisiting the Relationship between Hosts and Refugees in Kenya.” Global Movements for Refugee and Migrant Rights 21(3): 73–83. Bollig, Michael. 2000. “Staging Social Structures: Ritual and Social Organisation in an Egalitarian Society. The Pastoral Pokot of Northern Kenya.” Ethnos 65(3): 341–65. Crisp, Jeff. 2000. “A State of Insecurity: The Political Economy of Violence in Kenya’s Refugee Camps.” African Affairs 99(397): 601–32. Hendrickson, Dylan, Jeremy Armon, and Robin Mearns. 1998. “The Changing Nature of Conflict and Famine Vulnerability: The Case of Livestock Raiding in Turkana District, Kenya.” Disasters 22(3): 185–99. Hendrickson, Dylan, Robin Mearns, and Jeremy Armon. 1996. “Livestock Raiding Among the Pastoral Turkana of Kenya: Redistribution, Predation and the Links to Famine.” IDS Bulletin 27(3): 17–30. Horn, Rebecca. 2010. “Responses to Intimate Partner Violence in Kakuma Refugee Camp: Refugee Inter- actions with Agency Systems.” Social Science and Medicine 70(1): 160–68. Horst, Cindy. 2001. “Vital Links in Social Security: Somali Refugees in the Daadab Camps in Kenya.” New Issues in Refugee Research Working Paper 38, United Nations High Commissioner for Refugees (UNHCR). HRC (Human Rights Watch). 2015. “There Is No Time Left: Climate Change, Environmental Threats and Human Rights in Turkana County, Kenya,” October 15. https://www.hrw.org/report/2015/10/15/ there-no-time-left/climate-change-environmental-threats-and-human-rights-turkana. Jacobsen, Karen. 2005. The Economic Life of Refugees. Bloomfield, CT: Kumarian. Jansen, Bram. 2008. “Between Vulnerability and Assertiveness: Negotiating Resettlement in Kakuma Refugee Camp. Kenya.” African Affairs 107(429): 569–87. Keen, David. 1994. The Benefits of Famine. A Political Economy of Famine and Relief in Southwestern Sudan, 1983–1989. Princeton, NJ: Princeton University Press. 196 Appendixes KNBS (Kenya National Bureau of Statistics) and SID (Society for International Development). 2013. “Exploring Kenya’s Inequality: Pulling Apart or Pooling Together?” http://inequalities.sidint.net/ kenya/wp-content/uploads/sites/3/2013/10/Preliminary%20pages.pdf. Krätli, Saverio, and Jeremy Swift. 1999. “Understanding and Managing Pastoral Conflict in Kenya: A Lit- erature Review.” Report produced under DFID contract 98 6863, University of Sussex, United King- dom. http://www.eldis.org/fulltext/pastconf.pdf. Las sailly-Jacob, Veronique. 1993. “The Rural-Urban Interface in Africa: Expansion and Adaptation.” Canada Journal on Refugees 12(8). Mburu, Nene. 2002. “The Proliferation of Guns and Rustling in Karamoja and Turkana Districts: The Case for Appropriate Disarmament Strategies.” Journal of Peace, Conflict and Development 2(2): 10. McCabe, J. T. 1990. “Success and Failure: The Breakdown of Traditional Drought Coping Institutions Among the Pastoral Turkana of Kenya.” Journal of Asian and African Studies 25(3–4): 146–60. McCabe, J. T., R. Dyson-Hudson, J. Wienpahl. 1999. “Nomadic Movements.” In Turkana Herders of the Dry Savanna: Ecology and Biobehavioral Response of Nomads to an Uncertain Environment, edited by M. A. Little and P. W. Leslie, 109–21. New York: Oxford University Press. McEvoy, Claire, and Ryan Murray. 2008. “Gauging Fear and Insecurity: Perspectives on Armed Violence in Eastern Equatoria and Turkana North.” Small Arms Survey Working Paper 14, Graduate Institute of International and Development Studies, Geneva. Mkutu, Kennedy, and Gerald Wandera. 2014. “Policing the Periphery, Opportunities and Challenges for Kenya Police Reserves.” Small Arms Survey Working Paper, Graduate Institute of International and Development Studies, Geneva. Mwangi, C. W. 2012. “Women Refugees and Sexual Violence in Kakuma Camp, Kenya: Invisible Rights, Justice, Protracted Protection and Human Insecurity.” Research paper, International Institute of Social Studies (ISS). Newhouse, Leonie S. 2015. “More than Mere Survival: Violence, Humanitarian Governance, and Practical Material Politics in a Kenyan Refugee Camp.” Environment and Planning A 47: 2292–307. Ohta, Itaru. 2005. “Coexisting with Cultural ‘Others’: Social Relationships between the Turkana and the Refugees at Kakuma, Northwest Kenya.” Pastoralists and Their Neighbors in Asia and Africa 69: 227–39. Oka, R. C. 2011. “Unlikely Cities In The Desert: The Informal Economy As Causal Agent For Permanent ‘Urban’ Sustainability In Kakuma Refugee Camp, Kenya.” Urban Anthropology and Studies of Cultural Systems and World Economic Development, 223–62. ———. 2014. “Coping with the Refugee Wait: The Role of Consumption, Normalcy, and Dignity in Refu- gee Lives at Kakuma Refugee Camp, Kenya.” American Anthropologist 116(1). G. Background Paper: Implications of a Changing Social Order and 197 Structures for Peace and Development Økland, Malin Toftesund. 2014. “Being Young and Out of Place: An Explorative Study on the Everyday Practices of Educational Provision in the Humanitarian Operations in Kakuma, Northwestern Kenya.” Masters’ Thesis, University of Stavanger, Norway. ReDSS (Regional Durable Solutions Secretariat) and Samuel Hall. 2015. Devolution in Kenya: Oppor- tunity for Transitional Solutions for Refugees? Analysing the Impact of Devolution on Refugee Affairs in Refugee Hosting Counties. ReDSS and Samuel Hall. https://drc.dk/media/1419712/final_devolu- tion_report_230715.pdf. Sommers, Marc. 2002. “Peace Education and Refugee Youth.” In Learning for a Future: Refugee Educa- tion in Developing Countries, edited by Jeff Crisp, Christopher Talbot, and Daina B. Cipollone. United Nations High Commissioner for Refugees (UNHCR). http://www.unhcr.org/4a1d5ba36.pdf. Unruh, Jon D. 2005. “Changing Conflict Resolution Institutions in the Ethiopian Pastoral Commons: The Role of Armed Confrontation in Rule-making.” GeoJournal 64(3): 225–37. UNHCR 2015. (United Nations High Commissioner for Refugees), “Kenya: Kakuma Operational Update,” (October 15–28), UNHCR. Vemuru, V., R. Oka, R. Gengo, and L. Gettler. 2016. “Refugee Impacts on Turkana Hosts—A Social Impact Analysis for Kakuma, Turkana County, Kenya.” World Bank, Washington, DC. Whitaker, Beth Elise. 1999. “Changing Opportunities: Refugees and Host Communities in Western Tan- zania.” Centre for Documentation and Research, UNHCR. http://www.unhcr.org/3ae6a0c70.pdf. World Bank. 2014 “Kenya: Devolution Issues Note for the WB Country Partnership Strategy (FY14–18).” Implications and Potential Operations (draft), World Bank, Washington, DC. ———. 2015. Young People Democracy, and Governance in Kenya. 2015: What They Know and How They Engage. Washington, DC: World Bank. World Bank and DFID (Department for International Development). 2015. “Young People Democracy and Governance in Kenya. What They Know and How They Engage.” Formative Research and Draft Strategy. Well Told Story for the World Bank, World Bank, Washington, DC. H Background Paper: Opportunities for Sustainable Livelihoods Context, Rationale, and Methodology Turkana County has been hosting refugees from neighboring countries since 1992; its current population is an estimated at over 160,000—15 percent of the county’s popu- lation. An increase in the interaction between refugees and their local hosts has led to positive and negative outcomes, including employment opportunities for the host community, increased economic opportunities for both, intermarriage, and occasional confrontations between the communities over land or other issues. The impact of the presence of a large number of refugees in Turkana is both positive and negative. Their prolonged stay in the arid area of Kakuma has had many negative impacts on the local population, especially in terms of increased demand and cost of firewood and wood-burned charcoal, among other items. However, the Kakuma Refu- gee Camp has provided a large market for host community commodities and increased economic opportunities for traders in Kakuma town. The businesses operating in the refugee camp collectively produce an estimated US$350,000–400,000 in monthly sales (Oka 2014). Apart from anecdotal evidence, a systematic assessment, disaggregated by gender and age, of sustainable livelihood opportunities available for local host com- munities and refugees beyond pastoral livelihoods (for the host community) in farm-, nonfarm- and skills-based livelihood sectors, including entrepreneurship, has not been conducted. The scope and scale of these opportunities; the policy and regulatory frame- works; the major constraints to their adoption, including technical and entrepreneurial skills; the availability of finance; and the existence of local and regional markets and market information are not well understood. This paper specifically explores the potential positive contributions of the refugees’ skills and resources to the local economy as well as the increasing production capacity and consumption demand, which could stimulate an expansion of the host community economy. Focused attention is given to the state of livelihood opportunities in the host 199 200 Appendixes and refugee communities that have emerged from or that are related to the presence of the refugee camp and the intersection of refugee and host economies. Zones for inter- vention and growth that would benefit both communities are examined. This study builds on an extensive literature review, followed by an analysis of data on livelihoods currently being pursued by refugee and host community members, disag- gregated by gender and age group to discern trends, opportunities, and constraints. Qualitative intensive interviews and discussions were carried out with key informants comprising local host community and refugee members engaged in agricultural and pastoral livelihoods, entrepreneurs, aspirants, and business-owners, and those seeking skills-based jobs in the marketplace. The study looks broadly at the range of livelihoods being pursued by host and refugee communities, the scale of these activities, contribu- tions to household income, constraints to maintain current levels and/or scaling up, scope for diversification, understanding of the local refugee camp and regional input and output markets, and access to financial services and skills training. The Turkana People The Turkana people emerged as a distinct ethnic group sometime during the early to middle decades of the 19th century. According to history and oral tradition, the “origi- nal” Turkana were part of the Ateker language group of the eastern Nilotic linguistic family known as the Central Para-nilotes (Lamphear 1992). Oral history suggests that, prior to A.D. 1500, the ancestors of the Ateker language group lived somewhere in the southern Sudan and most likely subsisted as hunter-gatherers. At the beginning of the 18th century, the ancestral Ateker people began a southern migration and split into segments that were to form distinct linguistic groups, including the Karamajong, the Dodos, and the Toposa. Oral history suggests that a group called the Jie seceded from the Karamajong people and that the Turkana then separated from the Jie and expanded their territory by establishing themselves in the region near the Tarach River in what is now Turkana District. By the beginning of the 19th century, Turkana cattle camps began to push down the Tarach in search of new pastures for graz- ing their animals. As they moved westward, the Turkana encountered other pastoral groups, some of which herded camels (most likely the Rendille and Borana). As the Turkana expanded eastward, they began to assimilate and disperse other groups. They first pushed to the north and east to Lake Turkana, and then to the south, crossing the Turkwel River. They captured large numbers of animals during their expansion, includ- ing Boran Zebu cattle and many camels. H. Background Paper: Opportunities for Sustainable Livelihoods 201 From the 1850s onward, due to unfavorable climatic conditions Box H.1. The Myth of Atanayeche in the Turkana District that led to variable fodder and water supplies, and because of the unique requirements of each stock There is a myth among the Turkana peo- species, Turkana pastoralists developed a social system that was ple about Atanayeche—the mother of the flexible and a pastoral system that was well augmented with entire Ateker people, including the Tur- agriculture, hunting, gathering, and fishing (Lamphear 1992). kana. Atanayeche was a very hospitable The Turkana people had competitive raiding relationships with woman, so in times of drought, all of the surrounding pastoralist tribes. By the 1890s, prior to the first young men would leave their belongings arrival of the British military presence, the Turkana people had at her house while they went to look for extended their territory far beyond what they use today into green pastures in Uganda, South Sudan, South Sudan, Ethiopia, Uganda, and the contemporary Pokot and other parts of Kenya. She would and Samburu lands in Western and Central Kenya. gather, preserve, and then cook wild foods, and when the young men came The local Turkana have their own understanding of their ori- back hungry, she would feed them. She gins and their autochthonous relationship to Turkana land. taught the women how to gather and Oral historical data suggest that the Turkana of Kakuma as cook plants such as edapal, elamach, and well as northwestern Turkana believe their origins to be in the edu, and told them to not let this knowl- area of Lorugum, 50 kilometers (km) west of Lodwar on the edge die because it would keep the Ateker Turkwel River. The Turkana dialect is said to be spoken at its alive. Before Atanayeche died, she made purest in this area. It is also where, according to some, a leg- a big circle to indicate what would be the endary mother-hero figure called Atanayeche had settled (box Turkana homeland—her home—and told H.1). Others claim that Atanayeche had settled 50 km west of them to go forth. They went in all direc- Kakuma, near the Ugandan border. These ideas of autochthony tions to become the Toposa, the Jie, the shape the ties that the Turkana have toward their lands and serve Dodoth, the Karamoja, the Merile, the as a forum for resistance against external visitors, whether they Dongero, and the Nkoroma. be colonizers, invaders/raiders, or development/relief workers. They also serve as a powerful reminder to protect the Turkana lifestyle from outside interference. It is critical that actors engaged in development efforts keep in mind the Turkana perceptions of their origins and ties to their land and lifestyle when planning interventions and projects. Livelihoods of the Turkana People Livelihoods in Turkana are constrained by an arid environment; remoteness from the capital and other urban, industrialized, or agriculturally rich areas in the south; poor access to services; and other underlying causes of poverty also experienced elsewhere in Kenya. In Turkana County, 3 percent of residents with no formal education, 10 percent with a primary education, and 29 percent with a secondary or higher level of education are working for pay. The incidence of work for pay is highest in Nairobi at 202 Appendixes 49 percent—20 points above Turkana among those with secondary or higher level of education. The rest of the population depends on agropastoralism (20 percent), fishing (12 percent) and casual and formal labor (8 percent) in urban and periurban areas (FAO 2007). Crop production is carried out along the Kerio and Turkwel Rivers and on the arable flood plains; it is rain-fed and irrigated. The main cultivated crops are maize, sorghum, beans, and cassava. The Turkana practice nomadic pastoralism with livestock, including cattle, goats, cam- els, donkeys, and sheep. It is highly gendered activity—men and young boys are primar- ily concerned with herding and tending juvenile and adult animals. Women and young girls share in the pastoral labor by tending to and raising young animals from birth to the time that they can graze with the larger herds. In addition to being a primary source of livelihood, livestock constitute a central aspect of Turkana culture and social structure, functioning as sources of milk and meat as well as a form of currency for bride‐price negotiations and dowries. Grazing land is a critical asset for the Turkana people because they are pastoralists. In the Turkana County, the general nature of land tenure, as elsewhere among Kenyan pastoral communities, is that land is communally owned. Land rights are reserved to communities rather than to individuals, and land is an undivided part of the social system in which rightful use is determined by affinity, common residence, and social status. In Turkana, land tenure is a mixture of formal, customary, and community/group ownership. Formal tenure refers to land allocated by the state through a legal title. There is no standard size for formally allocated plots. In recent years, communities have had to diversify their livelihood activities to supple- ment pastoralism because it is insufficient in meeting the full range of economic and social needs. The bulk of this diversification has been undertaken by women who, along with helping to raise animals, are invariably trained in horticulture and the gather- ing and processing of wild plants and other foods, while continuing their practice in agriculture. Other diversification activities undertaken primarily by Turkana women include selling fish from Lake Turkana; weaving mats and baskets from the Doum Palm; collecting and selling aloe, gum arabic, honey, wild fruits, and firewood; and the production and sale of charcoal and alcohol (Watson and van Binsbergen 2008b). There is currently an increased emphasis on the processing and sale of skins and hides. Pastoralism and Pastoral Livelihoods According to Blench (2000), approximately 70 percent of the human population inhab- iting Turkana is nomadic or seminomadic and dependent on pastoralist livestock pro- duction for their survival. Five species of livestock are managed: cattle, camels, donkeys, goats, and sheep. Livestock, especially Zebu (humped) cattle, are very important in H. Background Paper: Opportunities for Sustainable Livelihoods 203 Turkana culture. Their value goes beyond the production of meat: it is based on the full set of services they supply: milk, meat, blood, and hides; their asset value as a form of savings used as traditional currency to negotiate for brides and dowry payment; and their cultural symbolism. Because of this, abandoning pastoralism—even if it becomes climatically, environmentally, or economically nonviable—would be difficult if not impossible for the Turkana people (ILRI 2006). Evidence from the field suggests that the Turkana see pastoralism as their special area of knowledge; assert that they have weathered various climatic, environmental, and economic crises before; and believe that their judicious use of pastoralism in conjunction with agriculture and foraged foods help them mitigate these types of crises. The traditional strategy of pastoralists is to move to areas with more rain and then return to traditionally drier areas when the rains arrive there so both pasture and foliage is renewed. Studies done in the 1980s and 1990s in north and south Turkana County suggest that the maximum movement of the Turkana ranged between 150–200 km in a circular route (McCabe et al. 1999). While the Turkana may have ranged further prior to and since colonialism, several factors continue to severely compromise the long-distance movement of livestock, such as the establishment of national frontiers, increased frequency of droughts, growing human and livestock numbers, insecurity, and encroachment into traditional dry season pasture by agropastoralism. In addi- tion, the sedentarization, rapid population growth, fragmentation, and privatization of formerly communally used pastures, the spread of wildlife conservation areas, and the increasing importance of agriculture are all factors that strain land resources, both within and across communities. While the tendency of pastoral households to accumulate livestock has been attrib- uted to cultural factors or common property tenure arrangements, there is growing awareness that herd accumulation is a rational self-insurance strategy to follow in an uncertain production environment (box H.2). Assuming that periodic herd die-offs are inevitable in this production environment and expected postcrisis herd size is an increasing function of precrisis herd size, herd accumulation is an effective risk man- agement strategy that pastoral households follow. Various cross-cultural studies among pastoralists suggest the presence of an asset (herd size) threshold above which the herd size is both resilient and robust across multiple crises/disasters, but below which herd sizes rapidly decline with each disaster (Murphy 2014). Ultimately, larger herds ex ante provide an effective, albeit costly, means of insuring sufficient herd size ex post (Watson and van Binsbergen 2008a). The acquisition or addition of livestock occurs through bridewealth, exchanges, gifts, payments, and begging—an important contributor to herd buildup in addition to births within the herd (de Vries et al. 2006). 204 Appendixes Coping strategies implemented by pastoralists in past years Box H.2. Building a Herd include slaughtering livestock and preserving the meat, pre- Traditionally, a young man is given a sin- serving grazing areas for times of extreme drought, dividing gle goat with which to start a herd, and large herds into smaller units and species, loaning stock among he will accumulate more through animal relatives and friends, collecting wild fruits and bartered cereals, husbandry. Once he has accumulated suf- and begging for food (ILRI 2006). The three main contributors ficient livestock, the animals are used to to huge drops in herd sizes are drought, disease, and dowry. negotiate for wives. Many Turkana men are polygamous, and livestock wealth Transformation of Pastoral Livelihoods determines the number of wives they can Droughts and conflicts have led to a transformation in the negotiate for and support. In Turkana cul- sociocultural and socioeconomic organization of pastoral soci- ture, men are in charge of livestock. They eties. The impact of drought is particularly acute for poorer make decisions concerning the animals members of communities with smaller livestock holdings and and control the financial capital. If a man less developed social support networks. The impacts include wishes to divide his herds into several humanitarian problems and localized degradation of natural groups, he can do it without discussing resources, which both contribute to long-term impoverish- the matter with his wife or wives. In the ment among pastoralists as many resort to selling animals at event that a married man dies, the fam- low prices and then are unable to restock after a drought ends ily livestock is passed down to his eldest (Notenbaert et al. 2007). son rather than his wives. If the son is too young to inherit the herd, a male relative With the loss of cattle and other livestock, women have started will care for it until the son is old enough. to play more active roles in ensuring household resilience by adopting a range of coping strategies, including increased engagement in diversified income-generating activities. At the same time, natural and man-made crises have led to an increase in female-headed households and has intensified women’s vulnerability due to their poor customary rights to access land, resources, and other assets (Omolo 2010). Livestock raiding among northern Kenya’s pastoralists has changed profoundly over the last few decades. Raiding has transformed from a quasi-cultural practice with impor- tant livelihood enhancing functions into a more targeted predatory activity. Predatory raiding occurs on a very large scale, is extremely violent, and is sponsored by actors with criminal motives from outside the pastoral sector (Hendrickson et al. 1998). Tur- kana herders, especially along the northern, western, and southern borders of Turkana County, are at constant risk of being violently dispossessed of their livestock. With their mobility already constrained by generalized insecurity, the impact on livelihoods is dev- astating. According to Grenier (2013), fought with modern weaponry and often with extreme violence, raiding is increasingly enmeshed in politicized claims over adminis- trative boundaries, struggles for exclusive access to land, and attempts to establish or H. Background Paper: Opportunities for Sustainable Livelihoods 205 safeguard an ethnically homogeneous electoral base. These conflicts are part of Kenya’s troubled politics of decentralization, and as such they must be viewed in the context of wider political developments in the country. Turkana County is characterized by persistent incidents of local-level conflict and vio- lence. Contestation over dry-season pastures has been a central focus of land-related disputes among transhumant populations and between villages, sublocations, and tribes.28 Climate change and a rise in the frequency of droughts have further contrib- uted to resource-based clashes between communities. The proliferation and availability of illegal firearms and pervasive sociocultural factors such as the warrior culture of moranism, in which young people are expected to raid livestock from neighboring com- munities in order to be qualified for marriage or praise, have further exacerbated violent conflict (ADB 2009). The actual rituals and behaviors regarding the preparation, prac- tice, blessing/sanction, and act of raiding, called ngingoroko, are a relatively recent part of Turkana culture, ideology, and cosmology, but this ritualized system of raiding has now been subsumed into a larger “cash market-oriented enterprise” (Eaton 2008). Some recent observations, including the current study, suggest that many of these raiding excursions are sponsored by members of the local trading community who supply cash and weapons in return for a cut or commission. Many of these “raided” livestock are taken over the border and sold in distant markets, a trend seen throughout this region (Hendrickson et al. 1998; Mulugeta and Hagmann 2008). One woman interviewed for the study says, “we women always know that there was a raid when we see a lot of cattle or livestock suddenly in the market.” Field data suggest that some nimuchurus (brokers) in the livestock market associations are also involved in predatory raiding, serving as gatekeepers into the markets and getting a commission for their participation. The Sale of Livestock The pastoralists of Turkana usually sell animals only when they must in order to meet urgent households needs, such as food (maize and maize flour), tobacco, bread, medi- cal and school fees, shoes, or clothing. Most of these cash needs can be covered with the sale of a few goats or sheep. Larger stock, particularly cattle, is hardly ever sold or marketed unless there is a dire need for cash. Camels are an important source of milk. Buying and breeding them is very expensive, and they have a high cultural value, so pastoralists avoid selling female camels at any cost (Watson and van Binsbergen 2008a). However, recent famines and droughts have made camel milk a premium commodity. One old man laments: 28. Other tribal communities include the Didinga, the Jie/Dodoth, and the Toposa from South Sudan; the Nyangatom who migrate between South Sudan and Ethiopia; and the Dassanech from Ethiopia. 206 Appendixes “But one food that has completely disappeared is edodo, dry milk, dry camel milk, that was [our] main source of local milk. The Somalis eat the camels, so the number of camels has gone down, and there is no edodo anywhere. You make edodo by taking camel milk, putting it in a big gourd until it separates, and then taking the curdled milk out to boil it. Once the curdled milk has boiled down, you put it outside to dry in the sun. Then you put it in a bag. When you want to cook, you take the powder, mix it with oil and milk. It expands and you eat it. It is very good.” (KII 2015) Sheep are occasionally slaughtered at home—mutton is considered a delicacy, employed as a “treatment” for human ailments, used for traditional rituals, and given as gifts. Sheep have another value: they are stores of fat. One young woman says, “If I could get 50 goats, no actually, 25 goats and 25 sheep. Even if drought comes and takes half, we still have 25–30 animals to rebuild the herd. Out of the 25 goats, I would have 5 male and 20 female goats, and out of 25 sheep, 5 rams and 20 ewes. The advantage of sheep is that if you slaughter sheep, you get a lot of fat that can be rendered and stored, it is used instead of oil, that you have to buy, for cooking, eating.” (KII 2015) Herders and livestock traders in northern Kenya, including Turkana, operate in unpre- dictable circumstances—risk and uncertainty is ubiquitous. Formal contract law to protect against business fraud does not exist. Trade is enabled by the existing network of relations based on kinship and ethnic values and by local institutions, such as the dalaal or middlemen. For the cattle traders of northern Kenya, insecurity is disguised in numerous ways and manifests itself in several forms, including ethnic conflicts, high- way robbery—also called banditry or shifta—and theft of livestock (Mahmoud 2008). Equal access to trading activities is limited because the entire import economy of Kakuma is controlled by 5–11 key players with contacts in Nairobi, Kitale, and beyond. As a cartel, it can strangle anyone who is not part of the network. The livestock markets are also controlled by brokers working together as a cartel inside and outside markets. They negotiate with the livestock holders and sellers, but in the market, they only sell to each other. While a big goat sells for K Sh 5,000 (US$50), they might only pay K Sh 200–500 (US$2–5) to the owner—a tenth of the actual price. The same equation would be applied to a cow worth K Sh 15,000–20,000 (US$150–200). The small producers and pastoralists are always disadvantaged, even if they are aware of the market price because they are often selling under duress for items such as medicine or food. Producers are, H. Background Paper: Opportunities for Sustainable Livelihoods 207 however, able to demand better prices if they travel to the Lokangai temporary market rather than the Kakuma and Lodwar markets because local and outside brokers com- pete there. Other Livelihoods There is anecdotal evidence of pastoralists starting businesses and owing lorries, rental houses, and hardware stores. Some drive boda-bodas (motorcycle taxis), and a few have found work in local restaurants in Kakuma town as cooks and cleaners. Due to low levels of educational achievement, most of the jobs available for locals are low-skill and low-wage, such as domestic servants, construction workers, watchmen, drivers, and unskilled laborers. However, because of the poor quality of schools and teaching, the youth continue to drop out of school at high rates. While this situation is undesirable, the communities around the camp have more options for both education and jobs and are likely better off than others in the county (KII 2015). Opportunities for basic education is very limited for the locals, and as a result, pastoralism remains the mainstay livelihood and safety net. But government efforts to sustain pastoralism are inadequate, including poor veterinary services, water points, and security from preda- tory raiding (KII 2015). In the words of one Turkana local: “We have always had cattle, goats, camels, and donkeys. We usually go by best number, and try to maintain that number. We sell some, get some, by bride- wealth or exchange, to replace. When we know we have to pay bridewealth, we try get as many as possible. Now we are more tied to market, we sell more and sometimes, now more, we buy calves and small goats.” (KII 2015) Some families collect small stones from the laga (dried riverbed) for construction, an extremely time-consuming activity. The stones are usually purchased by Somali builders. Livelihoods of Women Gender-differentiated roles between men and women delineate responsibilities for key livelihood activities. Men frequently engage in livestock trading, fishing, carpentry, construction work, long-distance hawking, and the provision of security services. They take advantage of comparatively more remunerative-waged employment than women (which often involves labor migration) and sell poles, rent buildings, and own shops (Watson and van Binsbergen 2008b). Women tend to engage in the petty trade of milk, 208 Appendixes porridge (uji), buns (mandazi), and vegetables; collect firewood, thornbushes, wooden poles, and wild fruit; process and sell fish, animal skins, charcoal, firewood, alcohol, and weaved mats and baskets; and offer their services to fetch water and do house- hold chores in the refugee camps. Women bear additional responsibilities for domestic tasks, including childbearing and raising, sourcing food, and caring for children and the homestead (ADB 2009). Women from areas near Lake Turkana, such as Kalokol, are engaged in wholesale fish businesses or sell dried fish fillets to markets in Lodwar; some also contract with hotels in Eliye Springs. Agriculture A large part of the administrative area in Turkana is rural. The dominant activity is subsistence farming, which is entirely dependent on the rainy season. Available data show that the rains are spread over very few days in the season. Returns on subsistence farming are poor due to very low land productivity, insufficient rainfall, poor soil qual- ity, and cultural issues. In Turkana, it is mainly the women who grow crops, including sorghum, maize, and vegetables, usually near the river. As one woman expressed: “We all learned agriculture from our mothers, sisters, aunts—it was like an apprenticeship” (KII 2015). Agricultural productivity was better in the past when there was plenty of water, plenty of forest, and fertile land. A Turkana woman explained: “When there is rain, I can get 12 bags of sorghum, 50 kg each, from 1 acre of land. I sell two out of these 12 bags in the market” (KII 2015). Traditionally, Turkana women have been agropasto- ralists, growing vegetables and sorghum and taking care of calves and baby goats. Some women have engaged in livestock rearing, but not pastoralism, including one woman in the Turkwel area west of Lodwar who maintained 20 cows, 500 goats, 15 camels, and five donkeys. Small Ruminants Local women employ a strategy to increase the size of a herd of goats “by either increas- ing the birth of new goats or by reducing the selling of the goats in the markets. Mostly, you should try to do both. During the wet season, the goats start giving birth, you have at least 10 to 20 animals, females giving birth to three to four small kids” (KII 2015). Selling of livestock is seen as a real problem that leads to the slow death of a herd, and it takes four to eight years to rebuild a herd (box H.3). In addition to firewood, goats represent liquid cash and are often sold during times of famine for food or medicine. Goats are usually sold in town through nimuchurus who approach livestock owners when they receive reliable information that large animals are being sold, but who let owners reach out to them for selling smaller animals. H. Background Paper: Opportunities for Sustainable Livelihoods 209 Charcoal Production and Sale Box H.3. The Rise and Fall of Herd Size Less than one percent of residents in Turkana County use liquefied petroleum gas (LPG), and “In 2015, my family had about 55 goats and 20 don- one percent use paraffin. Eighty-seven percent use keys. I remember in 1998, before I left, they had 50 firewood, and 11 percent use charcoal. Firewood is goats and 10 donkeys, and in 2000, they had 60 don- the also most common cooking fuel by gender—86 keys. When my father was there, the herd was grow- percent in male- and 89 percent in female-headed ing. In 2010, we had almost 200 goats. But because households. The level of firewood use is about 42 of bridewealth payments, and so many droughts, the percent, and the highest level of charcoal use is in herd size shrank again. It is like that, in the wet sea- Turkana County at 53 percent. Electricity use for sons, we grow the herd, and if the person leading is Lodwar Township Ward is 24 percent. good, and the women raise the kids (baby goats) well, then the herds grow. But the problem is always when Firewood collection and charcoal burning are key school fees have to be paid, or bridewealth payments sources of income and the largest business among need to be made.” (KII 2015). locals in Kakuma. The Ethiopians and Somalis buy charcoal; others buy firewood. Discussions with women reveal that: “Charcoal gives hotter flame and is more expensive but it also takes longer to prepare. So there is not that much difference between charcoal and firewood in terms of profit. Both are hand-to-mouth businesses, especially for the local women-mamas” (KII 2015). Locals either sell firewood or exchange firewood for maize, cereal, and oil. For example, firewood is sold in 10–15 sticks per bundle for K Sh 120–150, or it is exchanged for bowls of maize or cereal, which is preferred, because grains have a higher value than cash. One basin of charcoal sells for K Sh 150 in the camp and typically one family carries and sells about two basins of charcoal at K Sh 300 per week and earns about K Sh 1,200 per month. In place of money, locals are given three bowls of food for one basin of charcoal, which is worth K Sh 180–200 in the camp and K Sh 200–230 in town. Charcoal is mostly sold to the Somalis, followed by the Ethiopians. There has been a recent ban by the government on cutting trees based on environmen- tal concerns due to large-scale deforestation. However, charcoal and firewood is being brought from as far as Lopur, 25 km northwest of the camp; Lokang’ai, 25 km to the east; Namon, 70–80 km to the east; and Letea, 50 km to the west. The charcoal and fire- wood is transported by boda-bodas or piki-piki (motorcycle taxi), which has impacted their sale by women. 210 Appendixes Work in Refugee Camps According to the local women, the refugee camp is the most reliable way to get cash and food. Many local women are engaged in daily labor at the camp, such as washing dishes and clothes and carrying food for the refugees during distribution. In exchange for carrying around 26 kg of food, women are given 1–2 bowls of food, usually maize and (wheat) flour or sorghum, which can be sold for K Sh 100–200 (US$1–2); this option is preferred to wages, which would amount to K Sh 50 (US$0.50). Sometimes, women are given used clothes as wages. Some women wash clothes for wealthy refugees and earn K Sh 200–300 (US$2–3) per month. Many wash dishes, sweep the compounds, fetch water for the refugees, and carry food from the distribution centers in the refugee camps. Some mend fences: one fence-mending job can take a week to complete and fetches an income of K Sh 1,000 (US$10). The building of a new fence, including the collection of shrubs for the fence, carting it to the site, and building the fence, fetches an income of K Sh 3,000 (US$30). Small Business The refugee camp offers a market for milk, meat, firewood, and charcoal for the locals; it also offers a market for a range of products such as pots and pans, cooking utensils, bedsheets, mattresses, and even beds for the locals and Somali business people. Women are mostly involved in small businesses focused on food-related products (box H.4). The small activities undertaken by women include food kiosks, butcheries, milk kiosks, etc. These are considered culturally appropriate income-generating opportunities for women—within the culturally accepted roles and functions of women in the Turkana community (box H.5). The major constraint women face with expansion of current activities or setting up new enterprise relate to lack of micro-finance and credit facilities for women centered and oriented micro enterprises; in a context where women cannot provide collateral. Exploitative Livelihoods Poverty is abject among certain sections, and sexual exploitation of women and girls for food and money is not uncommon. There are jobs that locals find in the camp that are exploitative. Survival sex due to poverty is reported among single women with few options to feed their children, and some rape victims who earn a living from commer- cial sex. Begging, especially by older men and women, is reported in the camp. H. Background Paper: Opportunities for Sustainable Livelihoods 211 Box H.4. Profile of a Small Shop Owner A small shop owner in the local area sells oil, sugar, charcoal, and other consumable items. When she has the money, she uses K Sh 6,000 (US$60) to buy unga (maize flour), rice, spaghetti, and sugar from the Ethiopian wholesaler who gives her a good price—a kilo of sugar for K Sh 100 (US$1). She then packs these into smaller quanti- ties. A kilo of sugar is made into 15 packets each sold at K Sh 15 (US$0.15), providing a profit of K Sh 50 (US$0.50) per kilo of sugar sold. A similar approach is adopted for oil. Of the products bought, some is used for own consumption and the rest is sold. The sourcing of the different products is illustrative of the interconnected nature of markets. Typically sold items are sugar, oil, beans, maize, tea leaves, and salt (pur- chased from Ethiopians); charcoal (purchased locally); and baby formula, milk pow- der, sweets, matches, and other items (purchased in town from Somalis) (KII 2015). Amount/ Total Use Sale Price/ Total Total Item K Sh Per Unit Cost Home Shop Unit Sales Profit Sugar 8 kg/100 800 3 5 150 750 250 Oil 20 liters (125/liter) 2,500 10 10 150 1,500 250 Beans 5 (120/kg) 600 3 2 140 280 40 Maize 25 (28/kg) 700 15 10 35 350 70 Tea leaves 24 (60/dozen) 120 24 120 240 120 Salt 30 kg (6.7/kg) 200 5 25 10 250 33 Charcoal 3 bowls (150/bowl) 450 3 170 510 60 Total 5,520 4,240 1,033 The Children of Kakuma Although education is purportedly free in Kenya, a major constraint to educating chil- dren in Kakuma is the inability of many locals to afford the exam fees, uniforms, books, and food. Another important challenge is the unavailability of schools at the awi, forc- ing children to move to towns to attend them. Some children find work at the camp in the afternoon on school days and all day on weekends and holidays, earning around K Sh 1,500 (US$15) per month and sometimes getting fed. Children who obtain an education have opportunities such as joining the army, teaching, or working with the local Catholic diocese, as examples. 212 Appendixes Teachers also face limited support for their vocation. Box H.5. Profile of a Business Woman The resultant poor education outcomes include the My wife is a housewife right now. She was a busi- presence of young people on the streets who end up nesswoman. She sold small grocery items. Most doing odd jobs in the town and in the camp or who may things she sold, they were ordered from Kitale. She resort to “going the wrong way many times.” However, was a street vendor, in the Kakuma town market there is also a realization on the part of pastoralists of area. But she would not buy from the wholesal- the need for livelihood diversification within house- ers directly; she would buy from retailers here in holds, with some members continuing with herding town who would get things from the wholesalers, and the rest pursuing an education. who got them from Kitale. No small person buys directly from wholesalers. It was reported and observed that young boys and girls ages 10–17 work in the homes of refugees, fetch- There are two different types of buying for sell- ing water, cooking, sweeping, cleaning, or washing ing. She buys vegetables directly from retailers, clothes and dishes for 1,000–K Sh 1,500 (US$10–15) the Luhyas, and the Kikuyu. The Meru, she would per month. get tobacco from them. But grains and cereals, she would buy from refugee retailers, or the grain Emerging Livelihoods stores attached to the black market, the ones out- side the distribution points. But this has become The discoveries of oil and water and the devolution expensive. So she also went to the Kikuyus and process that is underway provide new avenues for live- the Luhyas, who have bigger groceries, and she lihoods among the Turkana people. can buy goods a little cheaper from them. But she Water always bought in small quantities, vegetables, fruit, because they spoil quickly. The refugees are the Two years ago, the Kenyan government announced that big market for our services and goods, charcoal, they found water in the Lotikipi plains in a vast under- boda-bodas, firewood, other things (KII 2015). ground lake the size of Lake Turkana. It “could provide water for Kenya for 70 years,” according to some experts (Avery 2014). This is seen as a real opportunity for the Turkana because the limited availability of water for irrigation has constrained the cul- tivation of land that otherwise has some potential.29 However, there are reports that the water is not fit for human consumption or for agricultural use due to high salinity levels. Locals are very suspicious of the reported results. And the idea that the discovery of water will bring a wealth of new opportunities to the local people may not be realistic because while private property is protected by the Kenyan constitution, water resources are considered national property. In this context and to allay fears among the local Tur- kana, careful management of the water resources is needed to ensure positive impacts 29. Intensive irrigation usually results in the accumulation of salt and minerals in the soil, which must be washed away every season, resulting in diminishing returns and ultimately collapse, according to studies about systems reliant on heavy irrigation. H. Background Paper: Opportunities for Sustainable Livelihoods 213 on the local community and a mutually favorable benefit-sharing arrangement. In addi- tion to the provision of water for irrigation, thorough attention to soil considerations in Turkana is critical because soil in arid lands is vulnerable to salinization, which destroys its agricultural potential. Further, while it may be possible to significantly increase the irrigated areas of Turkana County, massive financial investments would be required to set-up and maintain the irrigation system. Therefore, the potential for benefits accruing from the water in the near feature seems remote. Oil The main players regarding oil exploration in Turkana are currently Tullow PLC/Africa Oil and CEPSA. A planned oil pipeline and road network known as the LAPSSET corridor will traverse Turkana and link Kenya’s coast with South Sudan, Ethiopia, and Uganda, connecting it with a port at Lamu, Kenya, already under construction. There is suspicion among the local population that attention is being given to Turkana only because of the newly discovered oil reserves and that the Turkana people might not benefit from the oil revenues. Many blame rising insecurity on the oil, especially in southern Turkana, rather than on pastoralism or cattle raiding. The communal land that is held in trust for pastoralist communities by the county has been allocated to oil blocks and investment; and oil installations can displace pastoralists from important grazing sites and migratory routes. The current operations are probably already restrict- ing the livelihoods of the local population (Mkutu and Wandera 2015). Apprehension regarding the loss of water resources and environmental impacts is often voiced. Most locals are aware of the discovery of oil in the county, but many report high levels of exclusion: there are no Turkana in management positions for the companies han- dling the oil exploration contracts. They do not even work as clerks—only as watch- men, cleaners, and road marshals. Two years ago, hope and enthusiasm were evident at meetings regarding how the oil would be drilled; how it would be shared; and what other benefits it would offer to the local population, such as education and jobs. But the promised benefits have yet to accrue to the Turkana. While Tullow Oil, the agency contracted for oil exploration, is in place, there is real skepticism on the terms of the contract and who has and will benefit from the deals. For example, Tullow Oil provided education scholarships to the Turkana people that were channeled through political leaders and inequitably distributed. The lack of job opportunities in the industry has fuelled inter-clan conflicts among the Turkana, particularly as local politicians or elites secured opportunities for themselves. According to area police records, from January 2012 to April 2015, 14 demonstrations, road blockages, and attacks have occurred in Lokichar town. In October 2013, violent attacks on the company’s site forced managers to suspend operations for three weeks and to evacuate non-local staff members. Issues 214 Appendixes around oil and gas exploration have inflamed existing cross-border intercommunal rivalries and border disputes between the Turkana and the Toposa on the Kenya–South Sudan border and the internal Turkana–Pokot border. The Petroleum Bill of 2015 contains provisions about a community’s right to informa- tion and education and to compensation for displacement, for loss of a revenue source, and for environmental damage. It provides for a revenue-sharing formula of 20 percent to the county and 5 percent to locals, but the bill has still not passed. The locals who understand some of the bill’s provisions are more skeptical about whether or not any of the provisions will ever be translated into action or real benefits. The Land Act of 2012 provides for the designation of community land, such as that collectively owned by pastoralists and held in trust by the county. The Community Land Bill of 2015 provides detailed and complex steps for communities to register in order to secure their interests, but this process might be out of reach to many communities with low literacy rates and might benefit well-connected community elites. According to the Land Act, land may be forcibly acquired “in accordance with the law, for a public purpose, and upon prompt payment of just compensation to the person or persons, in full,” terms which are rather vague and which could be easily interpreted to suit the interests of the national government and investors. Compensation is not covered in any greater detail in the Community Land Bill, raising questions about who should be compensated (Mkutu and Wandera 2015). Devolution The Turkana County Integrated Plan prioritizes economic development programs that highlight key links between the refugee economy and aid delivery to potentially inte- grate refugee and host communities through sectoral priorities. Since the devolution process began in 2015, Turkana County has received the second highest share of the government’s budget (after Nairobi); most funds are spent on development projects. County authorities planning the economic development of Turkana recognize that refugees are an asset and see themselves becoming more active in the management of refugee affairs. They generally view refugees as having talents, resources, and connec- tions that the county can tap into for its own economic development (ReDSS and Sam- uel Hall 2015). Significant resources have improved infrastructure, notably for health and education, which the local population recognizes. Some affirm that devolution has led to benefits like increased jobs and medical and educational services. Doctors are being posted to interior places; schools are being built, including mobile schools to serve pastoral and nomadic communities; teachers are being employed, and famine relief is being provided under the Hunger Safety Net Programme (HSNP). Turkana has experienced an improvement in health indicators since devolution along with better H. Background Paper: Opportunities for Sustainable Livelihoods 215 equipment and functioning hospitals. Before devolution, the immunization rate among the Turkana was only 47 percent; it is now 72 percent. But everyone is not certain of the intentions or capacity of the county administration to effectively use the additional funds that are flowing due to devolution. Investments There was a boom in the local rental market and consequently in the construction sec- tor created by the devolution process in Turkana South and Lodwar, by the discovery of oil in Lokichar, and by the increased presence of refugees in Kakuma. Local families earning around K Sh 50,000 per month are able to save more than half that amount, which is then invested in buying land and building rental homes for outsiders working for various nongovernmental organizations (NGOs) and other service providers in the refugee camp. It is the outsiders who can afford these rental houses. Local land leasing and home and hotel room prices have reportedly risen rapidly. Some in the local population have benefited, but the poorer among them cannot afford these prices. The demand for meat has risen, boosting some pastoralists’ income; others are abandoning pastoral livelihoods or leaving school in pursuit of new opportunities. Cultural changes are noted, such as a cash economy over a cattle economy, increased alcohol consumption, and men and women staying away from their homes to work at the oil camps (Mkutu and Wandera 2015). Other investments include cable television, Internet access through smartphones, and sending children to good schools. Savings are used to buy beads that are used by women to increase their bridewealth: “a girl with a lot of beads will get more in bridewealth.” Others buy animals, mainly goats (a good size is 30–50 goats per awi), possibly sheep (10-15 because of the wool and the rendered fat), and some donkeys. Refugees at Kakuma Refugee Camp The Kakuma Refugee Camp, which hosts over 160,000 refugees,30 is one of the largest refugee camps in Africa. The refugee population is 15 percent of the county’s total pop- ulation.31 Established in 1992 for Sudanese refugees, the camp is now home to refugees from about 20 countries, particularly Ethiopians, Rwandans, Burundians, Congolese, Eritreans, Somalis, and Sudanese, including Darfurians. From an estimated population of 50,000 at the height of voluntary repatriation in January 2008, the camp population 30. According to data.unhcr.org, accessed September 2016. 31. See data.unhcr.org/SouthSudan. 216 Appendixes rose to over 170,000 by August 2014, with most families spending an average of ten years as camp residents. The ecosystem of the Kakuma Refugee Camp is described as a place where people of different nationalities and ethnic identification who speak multiple languages live (Ohta 2005). Their cultural and religious backgrounds also differ. The camp contains kindergarten, primary, and high schools; vocational schools; a hospital; health clin- ics; libraries; community centers; churches; and mosques. People engage in business at restaurants, general stores, and butcher and vegetable shops. In terms of income generation in the camp, despite the Kenyan government’s policy of encampment,32 2 percent of the population has access to in-camp employment as incen- tive workers, and 5 percent run their own businesses. Some of the population receives remittances from abroad, and 100 percent of the population depends on humanitarian aid. There are a number of constraints to the self-reliance of refugees, including limited employment opportunities and access to markets. Some refugees arrive with agricul- tural knowledge and business experience, among other skills, but lack opportunities to utilize their skills for livelihood purposes, contributing to their level of frustration. However, a number of vibrant economic and livelihood activities do exist that respond to the needs of camp residents. As part of this study, all of the stores in Kakuma town and Kakuma Refugee Camp were mapped; and in-depth interviews of camp traders and consumers were conducted, building on Oka’s previous work (2011, 2014). Kakuma 1 has four markets: Ethiopia, Mogadishu, Somalia Back, and Hong Kong; Kakuma 2 has two markets: Phase 1 and Phase 2; Kakuma 3 has three markets: Bantu, Darfuri, and New Market; and Kakuma 4, the newest section, has one market. Kakuma Refugee Camp has more than 2,150 shops, including 14 wholesalers. On the other hand, Kakuma town has 232 shops along the main road and in the alleys on either side of the road (figure H.1a–b). Buying from local shops on credit or borrowing from friends and neighbors to access food and other items is common among refugees. Food borrowed from relatives and neighbors is paid back in-kind after receiving rations from the World Food Programme (WFP) or in cash to local shopkeepers. Typical Dinka activities include the making and sale of beer out of sorghum (chang’ga) and the selling of tobacco. Most refugees 32. The Kenyan government implements an encampment policy mandating that refugees, once they have completed the status determination process, remain in a camp until more durable solutions are developed. This policy subsequently prevents refugees from owning cattle, cultivating, moving about freely, working, or integrating with local populations. As such, this policy has been criticized for rendering refugees entirely dependent on humanitarian support and for exacerbating divisions with local populations. H. Background Paper: Opportunities for Sustainable Livelihoods 217 Figure H.1. Total Businesses a. By Location 450 400 397 350 300 287 250 245 Percent 215 200 195 183 182 150 135 143 118 100 50 0 K1 K1 K1 K1 K2 K2 K3 K3 K3 K4 Mogadishu Somali Ethiopia Hong Phase 2 Phase 1 Darfuri Bantu New New Back Kong Market Market Market Area b. By Type 600 566 500 413 400 329 Percent 300 200 134 141 99 100 82 91 63 52 42 45 33 5 5 0 les ery re r are t ea it ic g to on er e es ran rbe rin on Fru an sto Oth o sho e/t sal ab tch rdw Ph /ph lo ch tau Ba get cs/ ffe ral Tai ng/ Bu Me Ha ics res Co eti ne Ve thi on / Ge sm tel Clo ctr Co Ho Ele Source: Vemuru et al. 2016. sell their food rations in the market to buy the goods they need. Refugees of various nationalities run their own shops, including the Somalis, Ethiopians, and Sudanese who own eateries; the Oromo who sell meat; and the Congolese, Burundese, and Rwandese who grow local vegetables in their backyards and sell them. 218 Appendixes Refugee women bear multiple responsibilities. In addition to caring for children, cooking, cleaning the house and compound, and washing clothes, women undertake income-generating activities, such as cooking and selling food (common within the camp) and managing small shops, and assume community roles, such as providing counseling services or participating in local politics. Older and younger women prepare food and sell it to restaurants and small hotels; some sell the food out of their houses. The most common type of food they prepare is the Ethiopian injera, chapati, and maandazi. Other income-generating activities include tailoring, domestic cleaning services, and fetching and selling firewood. Overall, women seek more diverse income-generating activities to supplement camp-provided assistance compared with men. While women have found alternative ways to earn incomes, men tend to express the sentiment that “it is difficult to find work.” Activities undertaken by refugee men include operating small- to medium-sized busi- nesses; working as drivers, traders, mechanics, and carpenters; and serving in commu- nity roles, such as block leaders and food advisory committee leaders. Some young men have jobs as incentive workers at the camp; others provide skilled labor as mechanics, electricians, and carpenters. Some young men are involved in the transport business as taxi or boda-boda drivers. Older men are more likely to be involved operating a grocery, meat, or stationery shop. Business opportunities available to men and women at the camp include milk trading; operating a grocery, clothing, or shoe store; running a telephone or electronic shop; or operating a restaurant or cafe. Movie theaters at the camp play movies and promotion videos of popular singers like Michael Jackson. Several restaurants have satellite broadcast dishes where residents can watch soccer games taking place across the globe. Business such as copy services, inter- national telephones, and international remittance services are open to serve to refugee population. Motorbike taxis buzz by on the bustling roads in the camp. Retailers of Kakuma Retailers in Kakuma operate from established shops and kiosks or, usually seasonally, in open air shops. Traders with established structures must obtain a license from the county council and pay the same annual fee required of Kenyan traders. The process usually takes only a few days. The allocation of space in the market by the market com- mittees poses one potential challenge in opening a new business. Open-air retailing is usually less regulated than in the established shops and kiosks. With the exception of livestock and to a lesser extent cereals and pulses in Kakuma, the role of wholesaling in the camp is limited. Most camp retailers are Somalis who benefit from a kin-based H. Background Paper: Opportunities for Sustainable Livelihoods 219 resource transfer system—the Somali hawala,33 which has roots in the “franco-valuta” system of the 1980s, in which local middlemen integrated into clan-based patrimonial networks exchanged foreign currency received from Somalis living abroad for shillings and delivered funds to their families (Sexsmith 2009). Retailers at the Kakuma camp get most of their cereal and pulse supplies from large traders at the camp or wholesalers in Kakuma town. The most significant supply source outside the camp consists of wholesalers in Kitale. Meru and Kikuyu merchants bring fresh produce in from Kitale, the central source for fruits and vegetables. Wholesalers located in Kakuma town are a secondary source for produce, especially if road condi- tions challenge the main supply from Kitale. Fruits and vegetables are often sold in shops, kiosks, and in open-air shops. The number of traders fluctuates according to the seasonal availability of certain produce and supply problems due to transport con- straints during the rainy season. There are several supply arrangements working in the camp: (1) producers sell directly to consumers in the camp; (2) retailers buy from producers in the outskirts of the town or camp and sell to consumers; (3) transporters serving as middlemen travel to nearby producing villages to collect milk and then either sell it directly to consumers or to open air retailers at the camp— the only form of wholesale trading in the milk market; and (4) women from the producing villages organize transport and make arrangements with retailers at the camp to receive the produce, with payments being made through the transporters (WFP 2014). Milk is sold at the camp and in town exclusively by female open-air retailers who do not sell any other products. The number of open-air retailers decreases during the dry season. In Kakuma, eggs are most commonly supplied from traders in Kitale to whole- salers in Kakuma town who subsequently source retailers in the camp; to a much lesser extent, they are also supplied by large-scale traders to retailers at the camp. They are sold in shops and kiosks as well as in open-air outlets by traders selling nothing else, except possibly packaged milk. Eggs are sometimes found in kiosks selling fruits and vegetables. The number of traders decreases during the rainy season. Fish is consumed mostly in its dry variety at the Kakuma camp. It is supplied to wholesalers and retailers in Kakuma town from Lake Turkana through Kalokol and from Lake Victoria through Kisumu and Kitale. Retailers in the camp get their supplies from the wholesalers in town. Fish is mostly sold in shops; kiosks; and, to a lesser extent, open air markets. The 33. The hawala system goes back a few thousand years and is practiced by a number of groups other than the Somalis, including the Lebanese, the Yemeni, the Sindhi, the Gujarati, the Marwari, the Bohri, the Khoja, and the Chinese. 220 Appendixes already significant number of retailers operating in the market increases during the dry season (WFP 2014). According to WFP (2014), key informants report that an important constraint to trade is the road conditions and associated high transport costs, especially between Kitale and Lodwar, but also between Kakuma town and the camp. Local traders and wholesal- ers use the Turkwel road via Lorugum into Uganda when the Kitale-Lodwar road is blocked. The wholesaler networks are affected when a seasonal river (laga) between Kakuma town and Kakuma 3 floods during the rainy season and transport is halted for one or two days, particularly affecting deliveries of perishable products. Other difficul- ties include refugee traders having to obtain travel permits to make purchases outside Kakuma (placing them at the mercy of the wholesalers). The movement restrictions are mentioned as an important trade deterrent. Other reported constraints to trade include a lack of trucks, poor access to credit, insecurity on the roads and at the camp, and high government taxes. Host–Refugee Community Interactions Interactions and economic activities between the refugee and host communities are expanding. Economic activities within Kakuma Refugee Camp and Kakuma town can be divided into eight categories (adapted from Oka 2011): n Formal relief redistribution. Food and other relief goods. n Formal commerce. Wholesale and retail traders supply goods and services to Tur- kana residents and to relief workers and agencies in Kakuma. n Formal-informal commerce. Wholesale traders supply goods and services to refu- gee wholesalers and retailers within Kakuma Refugee Camp. n Formal and informal banking and remittances. Four banks as well as mobile money systems enable cash and credit to be transferred from relatives outside Kakuma to the refugees. n Formal and informal employment. Refugees seek employment with relief agencies or the traders in Kakuma. H. Background Paper: Opportunities for Sustainable Livelihoods 221 n Informal overt commerce. Refugee wholesalers and retailers supply goods and ser- vices to refugee consumers within Kakuma Refugee Camp. n Informal relief commerce. Refugees sell portions of relief food packages to black market traders. n Informal covert commerce. Various actors are involved in the flow of drugs, weap- ons, and other extra-legal or illicit activities within Kakuma town and Kakuma Refu- gee Camp. The refugee camp provides a market for livestock, fish, firewood, and charcoal for host community traders. Refugee and host communities trade food items and other com- modities with one another. Remittances from family members and friends of refugees, including former refugees who have resettled in countries such as the United States, Canada, Australia, and Europe, increase the purchasing power of refugees in the camp and create a market for local businesses (Oka 2011, 2014). The camp acts as an impor- tant link between the refugee and the host population, with refugee traders providing a connection between the refugee camp economy and wider national and international markets. These links provide refugees with trading opportunities and exchanges with Kenyan producers and traders from the vicinity or from distant supply locations. To a great extent, Kakuma’s economy has been defined by the refugee economy, which has also contributed to the development of local economic architecture (ReDSS and Hall 2015) through: n Well-established economic networks. Strong business networks, predominantly run by Somali and to a lesser extent Ethiopian traders, recruit employees from within the camp and from their extended kin networks. They manage the flow of capital and credit within and between subcamps by offering small-scale retail credit for banking and remittances services. n Remittance economy/bringing cash to the local economy. Four banks within the Somali trader network have been identified through which remittance transfers are conducted, including Dahabshil, Amel, Dalson, and Iftin. n Black market and relief traders. The black market in Kakuma provides a mean for refugees to gain cash by selling their own goods and buying other goods and services. 222 Appendixes n Taxation and licensing issues. The deputy at the Department of Trade and Tourism confirmed that refugees are given licenses each year. Department officers provide this service in the camp as refugees cannot leave Kakuma. It is estimated that refu- gees in Kakuma contribute around K Sh 1.1 million in taxes per year. n A market for host community businesses. According to market assessments con- ducted by the World Food Programme (WFP), refugees spend an average of 60–70 percent of their disposable income on food and nonfood items provided by refugee and host community businesses. n Major source of employment for the local community. The presence of the refugee camp is also a major source of employment for the local community—from menial jobs provided by cleaners and housekeepers to local contractors who undertake con- struction of refugee facilities in the camp. n Purchase of livestock. Because of the high demand in Kakuma, livestock prices are higher there than in other parts of Turkana. Benefits from the Camp Refugee camp markets are used by the refugee and host communities alike. Most of the customers are refugees from the same or nearby subcamps, followed by members of the host community from nearby villages. The access to food from the refugee camp is reflected in nutritional data based on skinfold measurements (figure H.2.) that suggest that the host community of Kakuma has better nutritional access and status than that of Lokichoggio or Lorengo and similar access and status to that at Lorogum, which has benefitted from consistent development investments for over 20 years. For the most part, refugees shop at camp markets, but if commodities are unavailable there, they sometimes resort to the town market. Host community members make some of their purchases at the camp despite its distance from town because prices are lower. The price differential is attributed to the fact that most services are provided for free at the camp, which serves as a subsidy for retail businesses. Most of the traders in the camp are refugees, but the camp is not closed to local traders, and a limited number do operate there. However, the local community is emphatic that a closure of the camp would take with it most income sources and that the Kakuma town survives only because of the camp. Kakuma did exist as a town before the camp, and it is widely acknowledged that the H. Background Paper: Opportunities for Sustainable Livelihoods 223 camp has played a major role in the lives of the local people in terms of jobs and liveli- hoods. The presence of the camp has definitely brought in more trade, and the local population has benefited from the relatively lower prices at the camp shops. Those who work in refugee (predominantly Somali) households, are able to access small amounts of credit when needed. Figure H.2. Sum of Skinfolds among the Turkana (Mean Value, Age-Controlled) a. Women, n=300 20 25 30 35 Significant difference between Lorengo and Kakuma, Linear Prediction p < 0.001 Significant difference 15 between Lorengo and No significant difference Lorogum, p < 0.001 between Lorengo and Lokichoggio, p = 0.9 Kakuma Lokichoggio Lorengo Lorogum Location b. Men, n= 299 Significant difference between Lorengo and Kakuma, Linear Prediction p = 0.01 12 14 16 18 Significant difference between Lorengo and Lorogum, p = 0.06 No significant difference between Lorengo and Lokichoggio, p = 0.9 Kakuma Lokichoggio Lorengo Lorogum Location Source: Vemuru et al. 2016. 224 Appendixes Goods in Kakuma town are less expensive than in Lodwar, and even less expensive in the refugee camp. “People prefer to come to camp to buy goods because it is cheaper, sometimes even 30 to 40 shillings cheaper than in town.” “The shoes I bought in camp, just cost me 1000 shillings. In town, in Kakuma town, they will go for 2,000. I can get a kilogram of rice in camp for 90–100 shillings, but in town it will be 120 shillings. I make a profit of 20 shillings.” (KII 2015) The products available in the camp include solar panels at K Sh 7,500 as opposed to K Sh 9,000 in the town. Uneasy Relations There is an uneasy relationship between refugees and the local population. The sale of food by refugees in places where local people are begging for food is often cited as a source of ire among the Turkana. Mistreatment of refugees by locals is reported, such as when the local “mamas” (older Turkana women) who sell charcoal (makaa) at the camp compound are not compensated and are sometimes even beaten when asking for money—especially common when a large amount is owed. As one informant described: “While beatings are not too much, the refusal to pay is common” (KII 2015). Some local women are involved in survival sex and are at risk of physical and sexual violence. Camp officials can stop a local person from selling firewood and charcoal in the camp, but with the payment of a bribe—sometimes as small as 50 shillings—the official often allows the local person to sell their products. But this level of bribe would constitute an entire day’s profit for a “mama” or mzee (old man)—an enormous challenge that holds the locals back. By paying a bribe and cornering the market, the brokers are draining local livelihoods. There is disappointment among the host community regarding the limited opportu- nities offered by the camp in terms of jobs. The Turkana people who have low levels of education, status, and wealth and who are largely neglected by the government are only hired for menial jobs at the camp, such as watchmen and cleaners. Many have started businesses, combining nomadism and business. They interact with the market by selling goats and buying food. The constant struggle for basic services, food, and jobs contributes to the high stress levels among the Turkana. Although the presence of the camp has ensured the Turkana access to food, it is also a psychosocial stressor. Kakuma H. Background Paper: Opportunities for Sustainable Livelihoods 225 Figure H.3. Psychosocial Stressors Experienced by Turkana in Kakuma, Loki, Lorengo and Lorugum a. Kakuma b. Loki 1.5 1.5 1.0 1.0 Density Density 0.5 0.5 0 0 0 2 4 6 8 0 2 4 6 8 Total number of worries reported Total number of worries reported c. Lorengo d. Lorugum 1.5 1.5 1.0 1.0 Density Density 0.5 0.5 0 0 0 2 4 6 8 0 2 4 6 8 Total number of worries reported Total number of worries reported Source: Vemuru et al. 2016. demonstrates the greatest spread in the number of stressors reported, from 0 to 7 as seen from figure H.3. Conclusion The Turkana, a predominantly pastoral community, has witnessed transformation in livelihoods as well as economic organizations due to years of conflict and drought, and with the pastoral system being augmented over the years by agriculture, hunting, gathering, processing wild fruit, and fishing. The loss of cattle and other livestock has also redefined the role of women. While pastoralism continues to define the social and cultural identity for Turkana men, and cattle remain the go-to option to meet urgent household needs such as food (maize and maize flour), tobacco, bread, medical and school fees, shoes, and clothing. The Kakuma refugee camp offers alternate opportunities for employment and wage earning for both Turkana men and women. Low-end jobs for men include working as domestic servants, construction workers, watchmen, drivers, and unskilled labor; 226 Appendixes women are engaged in collecting firewood and burning charcoal, as well as providing daily labor in the camp, such as washing dishes and clothes and carrying distributed food for the refugees. There is anecdotal evidence about Turkana starting many small and some large businesses, but for the most part, they find it difficult to challenge the established cartels with better access to capital and networks. The entire import economy of Kakuma is thought to be controlled by 5–11 key players with contacts in Nairobi, Kitale, and beyond. A major constraint to accessing better jobs for the Turkana, especially those emerging from the discovery of oil and water or the process of devolution, is poor educational levels. However, the opportunities for basic education remain limited due to the poor quality schools and teaching in local public schools and high costs of exam fees, uni- forms, books, and food. It is not uncommon for Turkana children to find work in the camp, mainly fetching water, cooking, sweeping, cleaning, washing clothes, fetching water, and cleaning dishes. The refugee camp remains the major economic mainstay for the Turkana in Kakuma. Locals remain emphatic that the closure of camp could jeopardize most income sources and means of survival. However, concerted efforts are needed by the county authorities, development agencies, and nongovernmental organizations to ensure that: n Educational opportunities are enhanced for Turkana children by improving the quality of schools and teaching and, importantly, by making them both accessible and affordable for the locals. Culturally and socially appropriate options will need to be explored given the predominantly pastoral and dispersed population. n Youth need support with tailored programs for enhancing their technical skills along with training for life skills and basic and financial literacy so that they will be better equipped to seek employment in emergent areas. n Women’s livelihoods need to be supported, including improved access to input and output markets for agricultural and allied activities, as well as support, including capital and entrepreneurial training and inputs for establishing viable enterprises to augment family incomes. n Pastoral livelihoods remain central to the identity of Turkana men. Better support services, including veterinary care, are critical to enhancing resilience among the pastoralists. Support for livelihood diversification is also essential. H. Background Paper: Opportunities for Sustainable Livelihoods 227 References ADB (African Development Bank). 2009. “Socio-Economic Analysis and Public Consultation of Lake Turkana Communities.” Draft Report (December), ADB. Avery, S. 2014. “The Turkana Aquifer Discoveries and Development Proposals.” Discussion Brief (origi- nally produced for the Regional Learning and Advocacy Project for Vulnerable Dryland Communities (REGLAP). Blench, R. 2000. Extensive Pastoral Livestock Systems: Issues and Options for the Future. GCP/JPN/005/ JPN. FAO (Food and Agriculture Organization of the United Nations). Rome, Italy. de Vries D., P. W. Leslie, and Terrence McCabe J. 2006. “Livestock Acquisitions Dynamics in Nomadic Pastoralist Herd Demography: A Case Study Among Ngisonyoka Herders of South Turkana, Kenya. Human Ecology 34(1): 1–25. Eaton, D. 2008. “The Business of Peace: Raiding and Peace Work Along the Kenya-Uganda Border (Part I).” African Affairs 107/426: 89–110. FAO. 2007. As cited in Government of Kenya. 2015. Social Assessment Lesseru-Marich Passlodwar- Nakodok A1 Road South Sudan-EA Regional Transport, Trade And Development Facilitation Proj- ect (EARTTDFP).  Kenya National Highways Authority, Ministry of Transport and Infrastructures. Nairobi. Grenier, Clemens. 2013. Guns, Land, and Votes: Cattle Rustling and the Politics of Boundary (Re)making in Northern Kenya. African Affairs 112 (447), 216–237. Hendrickson D., J. Armon, and R. Mearns. 1998. The Changing Nature of Conflict and Famine Vulner- ability: The Case of Livestock Raiding in Turkana District, Kenya. Disasters 22(3): 185–99. ILRI (International Livestock Research Institute). 2006. Mapping Climate Vulnerability and Poverty in Africa: Where are the Hot Spots of Climate Change and Household Vulnerability? Lamphear, J. 1992. The Scattering Time: Turkana Responses to Colonial Rule. Oxford: Oxford University Press. Mahmoud, H. A. 2008. “Risky Trade, Resilient Traders: Trust and Livestock Marketing in Northern Kenya.” Africa: The Journal of the International African Institute 78(4). 561–81. McCabe, J. T., R. Dyson-Hudson, J. Wienpahl. 1999. “Nomadic Movements.” In Turkana Herders of the Dry Savanna: Ecology and Biobehavioral Response of Nomads to an Uncertain Environment, edited by M. A. Little and P. W. Leslie, 109–21. New York: Oxford University Press. Mkutu, K., and G. Wandera 2015. Conflict, Security and the Extractive Industries in Turkana, Kenya: Emerging Issues 2012–2015. United States International University—Africa and Kenya School of Government. 228 Appendixes Mulugeta, A., and T. Hagmann. 2008. “Governing Violence in the Pastoralist Space: Karrayu and State Notions of Cattle Raiding in the Ethiopian Awash Valley.” Afrika Focus 21(2): 71–87. Notenbaert, A. M. O., P. K. Thornton, and M. Herrero. 2007. “Livestock Development and Climate Change in Turkana District, Kenya.” Targeting and Innovation Discussion Paper 7, International Livestock Research Institute. https://cgspace.cgiar.org/bitstream/handle/10568/575/turkana_discus- sionpaper_7.pdf?sequence=1. Ohta, I. 2005. “Coexisting with Cultural “Others”: Social Relationships between the Turkana and the Refu- gees at Kakuma, Northwest Kenya. Pastoralists and Their Neighbors in Asia and Africa 69: 227–239 69. http://www.researchgate.net/publication/237630636. Oka, R. 2011. “Unlikely Cities In The Desert: The Informal Economy As Causal Agent For Permanent ‘Urban’ Sustainability In Kakuma Refugee Camp, Kenya.” Urban Anthropology and Studies of Cultural Systems and World Economic Development, 223–62. ———. 2014. “Coping with the Refugee Wait: The Role of Consumption, Normalcy, and Dignity in Refu- gee Lives at Kakuma Refugee Camp, Kenya.” American Anthropologist 116 (1): 1–15. Omolo, N. A., 2010. “Gender and Climate Change-Induced Conflict in Pastoral Communities: Case Study of Turkana in Northwestern Kenya.” African Journal on Conflict Resolution 10(2). ReDSS (Regional Durable Solutions Secretariat) and Samuel Hall. 2015. Devolution in Kenya: Opportunity for Transitional Solutions for Refugees? Analyzing the Impact of Devolution on Refugee Affairs in Refugee Hosting Counties. Nairobi: ReDSS and Samuel Hall. Sexsmith, K. 2009. Violent Conflict and Social Transformation: An Institutionalist Approach to the Role of Informal Economic Networks. European Journal of Development Research 21, 81–94. Watson D. J., and J. van Binsbergen J. 2008a. Livelihood Diversification Opportunities for Pastoralists in Turkana, Kenya. ILRI Research Report 5. Nairobi, Kenya: International Livestock Research Institute. ———. 2008b. Livestock Market Access And Opportunities in Turkana, Kenya. ILRI Research Report 3. Nairobi, Kenya: International Livestock Research Institute. WFP (World Food Program). 2014. Dadaab and Kakuma Refugee Camps Market Assessment. World Food Program. http://documents.wfp.org/stellent/groups/public/documents/ena/wfp266624.pdf?_ga=1.71 35026.1056326602.1473363188. Background Paper: I Socioeconomic Mapping and Analysis of Kakuma Introduction Established in 1992 in the marginalized county of Turkana in northwestern Kenya, Kakuma Refugee Camp is the epitome of a protracted displacement situation (Crisp 2002). Yet a closer look at the interactions between refugees and the host community surrounding the camp brings to light a range of dynamic economic interactions that point to an ironic situation where the host community has grown increasingly depen- dent on the camp for its own survival. Increasingly, the literature acknowledges the need to go beyond a representation of refugees as a passive group lacking economic agency (see for example, Betts et al. 2014), and Kakuma stands out as a particularly interesting case study. The common views of a refugee camp as economically isolated or as an economic burden to the host commu- nity do not explain the complexity of impacts that a vibrant community of more than 160,000 people34 has had on its surrounding environment for the past 24 years. Kakuma Refugee Camp was originally established for the Sudanese fleeing the conflict in their own neighboring country. It was set up very close to the town of Kakuma and has experienced succeeding influxes of refugees, including the arrival of Ethiopian refu- gees in 1994–95 and a significant group of Somalis from 1997 onward (Oka 2011: 229). The area has witnessed back-and-forth movements among Sudanese and South Suda- nese refugees following the chaotic evolution of the newly formed country. A recent example was in the spring and summer of 2014, when violence in South Sudan brought a sudden influx of about 40,000 South Sudanese refugees to Kakuma (UNHCR and WFP 2014: 29). The population of the camp has fluctuated, including a low in January 2008 of only 30,000 refugees (Oka 2011: 229). According to the latest data from the United Nations High Commissioner for Refugees (UNHCR), the current population at the camp is over 160,000. Two communities dominate in terms of their numbers: South 34. UNHCR Camp Population Statistics, September 2016. 229 230 Appendixes Sudanese, comprising about half the population (50.1 percent), and Somalis, compris- ing 30.3 percent.35 The World Bank’s social impact analysis, which includes this background paper, was designed to unpack the nuanced impacts of the refugee presence and provide recom- mendations to policy makers and developmental and relief agencies about how to foster positive socioeconomic interactions between the refugee and host communities. The objective is to go beyond the humanitarian canvas that currently frames interventions, including taking stock of the main factors that limit the socioeconomic development of Kakuma and its surroundings to assess what might be holding it back. This analysis seeks to verify a series of factors—some structural and related to the county’s history of marginalization, and some more directly related to the nature of the relationship between the refugees and hosts—that limit socioeconomic opportunities in Kakuma. These factors are reviewed based on qualitative and quantitative data collected dur- ing May–June 2015 for this research in addition to the existing literature. The data set includes a range of in-depth interviews and focus group discussions conducted with refugees and members of the surrounding host community in Kakuma, Lodwar, Lorogum, and Lokichoggio in Turkana County. The study also relies on quantitative data from a psychosocial stress and health survey of the host community surrounding the refugee camp. Structural Factors Shaping the Social Economy of Kakuma The story of the refugees at Kakuma and—perhaps even more—of the Turkana whose community surrounds the camp is of hardship and survival, including severe food inse- curity, irregular incomes, fragile livelihoods, and various forms of intra- and intercom- munal violence. These sentiments were clearly expressed during in-depth interviews conducted with members of these communities. Aridity, Drought, and Survival Several structural factors are important to consider when looking at the current socio- economic dynamics at play in Kakuma and in the county of Turkana more broadly. The county is located in an arid and semiarid area of Kenya, characterized by rainfall levels of 200–500 millimeters per year, irregular rain patterns, and frequent droughts (IFAD 2000). Reid and Ellis (1995) showed that Turkana County experienced multiyear droughts every five years on average. Episodes of multiyear drought are particularly lethal for livestock, which can usually only survive a one-year drought. Many of the Turkana 35. UNHCR Camp Population Statistics, August 2015. I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 231 interviewed for the present study have vivid memories of severe droughts over the past decades that decimated their herds and brought them to the verge of famine. In an environment mostly unfit for cultivation due to its short growing periods, local Tur- kana communities are composed of pastoralist tribes that rely on mobility to ensure that their livestock can access grazing lands during dry seasons. Most studies estimate that approximately 70 percent of the Turkana population is nomadic or seminomadic (Watson and van Binsbergen 2008). As Hendrickson et al. notes: “The Turkana, like the majority of pastoralists in Africa, have traditionally led a lifestyle geared towards subsistence. Though precarious at the best of times, it is well suited to the harsh, dryland environment they inhabit. Drought and famine are constant hazards….” (Hendrickson et al. 1996: 18) While the Turkana pastoralist lifestyle has always been vulnerable to severe weather variations, a common theme emerges from the literature and interviews about the envi- ronment becoming increasingly harsh over the last few decades due to more frequent droughts and decreasing average rainfall levels. Issues surrounding the direct impacts of climate change on an environment like Turkana are extremely complex and require nuanced assessment, but several trends that directly impact the county’s environment are worth noting because they will make it more difficult for the population to sustain- ably adjust: n Increase in the county’s population. The population of the county has significantly increased over the past decade, and county authorities estimate that the current annual growth rate is 6.4 percent, according to a recent Human Rights Watch report (HRW 2015: 49). There are currently more than 1.2 million inhabitants in the county, a significant increase from the 165,000 population reported in 1969 (Hendrickson et al. 1996). The density of the county’s population is still quite low, but its growth has put pressure on the county’s very limited resources and has degraded the environ- ment, further threatening the sustainability of pastoralist livelihoods. n Land degradation and concentration of livestock: “Unsustainable human activities in already fragile areas of northern Kenyan [arid and semiarid areas] have aggravated the conditions arising from natural disturbance such as drought or flooding leading to land deg- radation and desertification. … In Kenya, 78 percent of the land is threat- ened by desertification.” (Pragya n.d.: 33) 232 Appendixes The increase in the population and the high levels of violence among pastoralists, including the practice of cattle raiding, have led to a contraction in the amount of grazing land accessible to Turkana tribes. The violence has also led to a degrada- tion of traditional grazing lands and water points in areas no longer in use due to insecurity. At the same time, the concentration of livestock in safe areas has caused ecological degradation there (Pragya n.d.: 10). n Development projects impacting Lake Turkana on the Ethiopian side of the Omo River. As Avery studies in detail, the hydrological consequences of the man-made projects up the Omo River, which feeds Lake Turkana, are significant. The researcher estimates the current project will “require over 30 per cent of the Omo flow as a minimum … the potential reduction of inflow to Lake Turkana is therefore huge, and far greater than previously reported” (Avery 2013: 34). Furthermore, the Turkana’s pastoralist lifestyle has been challenged for decades by conflict, intertribal violence, and cattle raiding, which have decreased their capacity to adjust to drought (Avery 2013: 18). Several authors note that cattle raiding evolved from a cultural practice to a more predatory form, characterized by increased levels of violence and criminal activity. This change has further destabilized the traditional life- style of the Turkana who, as their mobility was increasingly constrained, experienced a significant impact on their livelihoods (see for example, Hendrickson et al. 1998). Aridity makes agricultural production difficult and irrigation necessary in the county. And Turkana men do not consider it proper to engage in agriculture, posing a cultural barrier to the practice. Turkana women, on the other hand, show a refined knowledge of their environment. Skills are passed from one generation of women to the next, and agri- cultural products provide a welcome additional income or in-kind revenue for Turkana families. All of the female respondents interviewed on the subject noted that the absence of water and aridity were the main obstacles to increasing their production capability. The combined impact of land degradation, insecurity, and repeated droughts leaves Turkana pastoralist populations on the very edge of survival. The living conditions in this difficult environment have been further complicated by decades of political mar- ginalization by national authorities. The County’s Historical Marginalization To understand the current situation of Kakuma, it is critical to appreciate Turkana County’s history of marginalization. The county has long been neglected by the Kenyan government, which considers it to be remote and backward due to its pastoralist I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 233 traditions. Most interactions between the Turkana and the state have involved efforts that clash with the pastoralist lifestyle. Two vivid examples in the collective memory of the Turkana are the campaign of forced schooling for children and the general disar- mament of Turkana tribes. The opposition to pastoralism by the government persists: authorities consider pastoralism to be an obstacle to development. Interviews with authorities conducted for a study on the impact of the devolution process on Kakuma illustrate the point: “Pastoralist lifestyle is a challenge: trying to improve the lifestyle of pastoralists is very difficult as they keep migrating. Trying to improve services for them is very difficult.”— Ministry of Devolution and Planning (Samuel Hall 2015: 18) General neglect by the central government has seriously impacted Turkana’s develop- ment. First, it has translated into very poor infrastructure: “Mandated support for such essential infrastructure as holding grounds, watering facilities, roads … is Spartan, if not completely absent. For example, the only good road in Turkana is the one between Lodwar and Lokichoggio.” (Watson and van Binsbergen 2008) The ongoing devolution process should help advance development in the county, but significant efforts will be needed to fill the development gap that decades of neglect produced. Access to basic public services, including health, education, and road infra- structure is quite poor among the Turkana; and the level of education in the county is low. According to the Kenyan National Board of Statistics, “only 3 percent of Turkana County residents have a secondary education … as many as 82 percent of Turkana County residents have no formal education.” (KNBS 2013). Another study focused on analyzing the socioeconomic context of areas surrounding Lake Turkana noted: “the average distance to a health facility in Northern Kenya is 52 kilometers, compared to a national norm of 5 km” (quoted by Avery 2012: 85). It is not surprising that, according to the government of Kenya, Turkana is the least developed of all Kenyan counties. Based on 2005–06 national statistics, 92.9 percent of the Turkana population is considered poor, a striking figure when compared to the then-national poverty rate of 46.6 percent.36 The marginalized situation of the county population compared to the rest of Kenya remains troubling: the Human Poverty Index of Turkana is 0.613 compared with a national average of 0.291 (Samuel Hall 2015: 36). 36. According to www.opendata.go.ke (October 25, 2015). 234 Appendixes This low level of development and the harsh environment have very clear consequences on the malnutrition levels of a population always on the edge of survival. A psychoso- cial and health survey conducted among the Turkana population found that a majority was underweight, with a body mass index BMI of less than 18.5, and that 24 percent of the respondents suffered from moderate or acute malnutrition (BMI < 17). While this survey did not include children, past studies found that malnutrition was com- mon among Turkana children as well.37 Malnutrition has a proven long-term effect on development. As noted by the World Bank, “... the effect of under-nutrition on young children (ages 0–8) can be devas- tating and enduring. It can impede behavioral and cognitive development, educability and reproductive health, thereby undermining future work pro- ductivity.” (World Bank 2015) Local Governance The incapacity of local authorities to respond to the needs of the Turkana population is both a cause and consequence of the county’s history of marginalization. A general feeling of defiance among the Turkana toward the national authorities in Nairobi and toward their own local leaders was palpable during the qualitative interviews conducted for this assessment: “People who need shelter should be received as guests, and I am glad that the UNHCR is doing things for them, but we are angry at our government, local leaders, and the big politicians in Nairobi, who treat us like garbage and have not paid any attention to us.” —Teacher and member of the Turkana community (KII 2015) This quote captures a general and somewhat common sense of frustration toward local and national elites, but qualitative data confirm that this feeling is deeply grounded and born out of the daily experience of Turkana communities. For example, water manage- ment is a key issue for the host community, whose members pool money to pay for the maintenance of its water pumps, but local dysfunction directly impact their lives and subsistence: “We have to pay 300 shillings per month for maintenance. We always pay to a committee but the committee, they eat the money. The pumps only last two years and then they get damaged.” (KII 2015) 37. See, for example, Childhood Disability and Malnutrition in Turkana Kenya on International Centre for Evidence in Dis- ability website at http://disabilitycentre.lshtm.ac.uk/files/2014/10/Malnutrition-and-Disability-Report-Full-Report.pdf. I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 235 Kakuma’s Economy and the Impact of the Camp on Turkana Communities In this context of historical marginalization, the establishment of the Kakuma Refugee Camp—ironically represented a lifeline that many in the Turkana community were ready to accept. The qualitative data collected for this study clearly shows that many Turkana households have reorganized their livelihoods around the presence of the camp. The camp created the environment for interesting—if strikingly unbalanced— economic relations to emerge between refugee and host communities. Legal Framework for Refugee Management in Kenya A review of the impact of devolution on Kakuma and Dadaab noted that the recent movement toward devolution does not directly impact refugee management because that responsibility is increasingly being centralized rather than devolved (Samuel Hall 2014). Refugee management remains at the national level under the portfolio of the Department for Refugee Affairs. The 2006 Refugee Act is the main legal instrument for refugee management, even though most stakeholders, including the government of Kenya, agree that it does not comprehensively cover the issue nor does it adequately protect refugees. A new bill was drafted by the government, working with civil society partners, but it has not yet passed into law. Because the management of refugees is being increasingly securitized, the Department of Refugee Affairs has been reinforced as a central institution for refugees, confirming the national ownership on the issue. County authorities have only peripheral responsibilities with regard to refugees, usually linked to sectors under their direct responsibility, particularly the provision of public ser- vices—such as health, road infrastructure, and education. Despite the nationalization of the refugee issue, counties will likely gain an increasing voice at the table, especially as they are gaining control over a sensitive issue for refugees—land (Samuel Hall 2014: 8). The practical consequences of this evolving legal framework are “the lack of coherent provisions for refugee management and protection” (Samuel Hall 2014: 11) in a context where at least five key legal texts partly or exclusively address the issue of refugees. In practice, however, Kenya still implements an encampment policy that does not grant freedom of movement to refugees and does not give them access to employment out- side the refugee camp. Kenya does not recognize local integration as one of the durable solutions available for refugees living on its soil even though much of the population already lives in and is integrated into the economic life of cities, including Nairobi. The practical consequences at the camp level are that refugees are restricted in their movements outside the camp: they need authorization to leave and a curfew forbids any person who is not a refugee to be inside the camp after 6:00 p.m. 236 Appendixes Figure I.1. Number of Businesses Per Type of Services/Goods Other (charcoal, firewood, enjeera, etc.) General store (various) Clothing and other Fruit/vegetables Hotel, restaurant, bar Butchery/fish Electronics, phone, Mpesa Tailoring Hardware Barber Cosmetics and other Coffee/tea Mechanics, garage Cyber café 0 100 200 300 400 500 600 700 Source: Vemuru et al. 2016. Inside the Camp—A Relief-Oriented but Dynamic Economy Past research has highlighted the vitality of the camp economy. Pérouse de Montclos and Kagwanja (2000) describe Kakuma refugee camp as “a market town,” adding that the “trading networks of the camps supply the surrounding areas as far away as the refu- gees’ countries of origin. … Each community has its own patterns of trade.” Qualitative interviews showed that host communities had noticed the impact that various refugee communities had on the overall economy of the camp and its surroundings, pointing at the Somali and Ethiopian communities as particularly active. As part of the present study, the research team conducted an assessment of the refugee camp’s businesses and found 10 medium- to large-sized markets in the camp and a total of 2,382 businesses located in the camp or in Kakuma town. Many businesses specialize in trading key goods and services (figure I.1). This assessment reveals a variety of businesses and economic activities that characterize Kakuma Refugee Camp. Patterns of trade are based on and complement the free aid received by the refugees, as studied in detail by Oka in his 2011 paper that showed the extent to which the refugees have operationalized the aid they receive and make use of to buy goods and commodities that they cannot access through the relief food package, thereby sustaining various types of businesses at the camp (Oka 2011). Oka noted that many refugees sold part or all of their food rations to get cash to buy other goods. By doing so, refugees reaffirm a sense of agency over their household patterns of consump- tion despite the general system of relief and assistance. In addition to selling relief goods I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 237 as sources of cash, refugees rely on remittances and, to a lesser extent, salaries from employment: “Minimal estimates of cash inflow from employment (66,000 USD), remit- tances (200,000 USD), and the black market (89,000) total 355,000 USD per month that circulates through and out of the refugee camp through the camp traders.” (Oka 2011: 249) Positive Economic Impact for Host Communities The socioeconomic consequences of the presence of the refugee camp on the lives of the host community are diverse, but several respondents note that the camp is one of the main livelihood providers for the Turkana population. Others note that their survival is inextricably linked to the camp, pointing to the ambiguous nature of the relationship between the two communities. Charity and Survival While Turkana communities usually depend on cattle for their survival, many respon- dents note that Kakuma Refugee Camp now represents an important component of their survival strategy. Charity and begging in the camp are common activities for Tur- kana living around it. A 28-year-old mother of seven children notes that due to her lack of education, begging or asking refugees for charity is one of the few coping strategies available to her when the situation is critical and she cannot feed her children. Several Turkana working in the camp claim they have established relationships with refugees, so positive that they would receive small gifts and support, usually in the form of food surpluses or a provided meal. Cheaper Goods Available Another important consequence of the camp’s presence on the daily lives of the Tur- kana is the availability of less costly goods. Turkana communities use the camp to pur- chase necessary items at cheaper prices. The availability of free food and other needed commodities drives prices down at the camp, to the benefit of the Turkana. The host community also benefits from the availability of a broad range of products imported by refugee traders and their business partners. A package of flour that sells for K Sh 40 (US$0.40) in the camp would reportedly cost K Sh 60–80 (US$0.60-0.80) in town; and one kilogram of rice would cost K Sh 90 (US$0.90) in the camp or K Sh 120 (US$1.20) in town. Luxury goods are accessible in the camp: one Turkana man reports having just purchased a pair of shoes for K Sh 1,000 (US$10)—double the estimated price in town. 238 Appendixes He also explains that solar panels were available in the camp for K Sh 7,500 (US$75) as opposed to K Sh 9,000 (US$90) in town. Menial Jobs The camp provides a number of job opportunities for Turkana men and women, mostly unskilled, irregular, and low-paying, but they do represent an important source of income for the host community. Male and female Turkana often combine a variety of menial economic activities—inside and outside the camp—to meet their family’s needs. Activities vary based on the demands in the camp, but Turkana women engage in activities such as carrying food for refugees, fetching water, washing dishes and clothes, mending fences, and collecting and selling firewood or charcoal. Some activities bear significant risks, especially sex work—called “survival sex” in the camp, which brings with it a high risk of HIV transmission and gender-based violence. Turkana children often work as servants in the camp, either in addition to attending school or as a full- time activity. Men find irregular employment in construction or in transporting material into the camp. Several Turkana note that they could not access skilled labor opportunities in the camp, mostly of which are provided by nongovernmental organizations (NGOs) and international organizations, either because their level of skills and education were too low or because the positions were awarded through nepotism or bribery, which blocked their access to these lucrative opportunities. “I went two days ago to the camp, and made a fence. For that, I got 1,000 shil- lings in cash. I also carried food for people, for a family of three. That was 50 shillings per person, so 150 shillings and two bowls of food. So I came back with 1,150 shillings and two bowls of maize. Two days before that, I got 1,000 shillings for mending four fences. Mending and making fences is seasonal, and there are so many people to build fences that they cannot hike prices.” —35-year-old Turkana woman (KII 2015) The refugee camp therefore has a direct impact on the Turkana’s access to food, income, and improved nutritional status, as demonstrated by the nutrition data collected for this study. This survey provided sum of skinfolds measurements, which demonstrated that the Turkana living in Kakuma have a better nutritional status than the Turkana of Lokichoggio or Lorengo. I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 239 Limited Access to Financial Services for Both Communities In the Camp There are only limited options for refugees to access formal banking services, including loans and credit. Equity Bank opened a branch in Kakuma, which offers schemes of revolving loans for a small number of refugees. Yet a less formal banking system is active in the camp that fills in the gaps of the formal system. To access credit and funds, some of the refugees can use Somali and Ethiopian traditional systems of hawala. These services are usually only accessible to members of these communities, which excludes the rest of the camp population. Four main hawala providers operate within the camp’s Somali business community: Dahabshil, Amel, Dalsan, and Iftin (Oka 2011: 244). This system allows refugees to access signifi- cant amounts in remittances, either from the diaspora or from relatives in the region. Oka estimates that the total international remittances reaching the camp amount to US$100,000–150,000 per month (Oka 2011). The other main channel of funds acces- sible for refugees is M-Pesa, a mobile-based system for transferring money. Refugees can use M-Pesa to receive remittances or to conduct transactions within the camp. A similar amount is sent to the camp through M-Pesa: about US$200,000 per month (Oka 2011: 245). Outside the Camp Local community access to formal banking and credit is also limited and actually partly depends on the refugee presence. In focus group discussions, Turkana woman expressed their defiance toward the formal banking system as a savings institution: “A goat is our bank. … Banks are not controlled by us and not by God. Men run banks, down-Kenyans run banks. If we give money to them and they disappear, what shall we do?” “The problem is that we are afraid of banks and the people who work there. We get afraid of making mistakes. But we have bank accounts. When we get HSNP [Hunger Safety Net Programme] money, we put it in there. We also have M-Pesa but that is not savings.” (KII 2015) Formal banks are perceived as institutions run by outsiders, and low financial literacy levels among the Turkana makes some uncomfortable with using the formal banking system. 240 Appendixes The Turkana seek small amounts of credit to buy food and other goods from the Somali shops in the camp or sometimes from the shops in town. Their level of access varies and is limited to the members of the community known for having reliable sources of income, such as a full-time job with an NGO in the camp. “Somalis sell goods cheaper, they know customer relations, they also give credit. If you work for an NGO and they know you, you can even get up to 100,000 shillings credit. As long as you keep paying back.” —26-year-old Turkana man (KII 2015) The Turkana, however, are often forced into giving credit to their refugee clients at the camp. Many report being asked for their goods on credit, particularly firewood and charcoal, and then must chase down their clients for payment, which often sparks ten- sions and disputes. This illustrates another facet of the unbalanced relationship between the refugee and host communities. 3.5 Structure of Investments in Both Communities Refugee Communities The investment structures of the Turkana and refugee communities are very different and shaped by the socioeconomic context in which they both developed. As noted above, refugees have invested significantly in the business architecture of the camp, filling the gaps of development assistance with a large range of services and distribution of goods. The camp’s Ethiopian and Somali communities have been particularly active in devel- oping businesses as a form of short- and long-term investment characteristic of the transient situation of refugees. Somalis and Ethiopians established long-term wholesale and other large businesses requiring a significant investment in permanent construc- tion, sometimes with the support of local Kenyan Meru and Kikuyu communities. These investments respond to long- and short-term strategies because most refugees plan to sell their businesses when they leave the camp, and a major objective of starting a business is to have cash for resettlement. Equally important, most refugees do not expect to leave in the immediate future and are therefore constantly reinvesting in their businesses, making them simultaneously think in both the short- and long-term. The Ethiopian and Somali communities at the refugee camp invest in their countries of origin in addition to their proposed places of settlement, such as the United States, the I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 241 United Kingdom, and Australia. Balancing between short-term strategies at the camp and the longer-term objective of resettlement is characteristic of the sense of “transient permanence” or “static transience” among refugees stuck in the limbo of protracted displacement (Oka 2011: 233). Meanwhile, they actively contribute to the overall devel- opment of the camp and its transition into a quasi-town. The Turkana Community For the most part, the Turkana engage in strategies of survival and subsistence, which does not leave much room to make investments. A socioeconomic survey conducted for this study showed that 45 percent of Turkana respondents worried first and fore- most about hunger. The Turkana economy has only recently become cash-based, and some business interactions with refugees are still made in kind, further limiting avail- able investment options. Qualitative interviews show that beyond mere survival, Turkana “invest” in three main ways, expecting a different type of return for each and with varying levels of confidence in their strategies, as outlined below. 1. Education and School Fees Second to food, education is a major expense for the Turkana. These pastoralist families noted that needing to pay school fees (or a bride-price) was one of the few reasons a household would agree to sell some of its livestock. Turkana children can receive free education in public schools, but the expenses related to education—such as books, transportation, and uniforms—are not negligible for these very low-income households. Qualitative interviews confirmed the significance of education expenses in Turkana household budgets: “We have 25 goats. We had more, around 60, but many have gone to pay school fees for children. … If the people, the big ones, keep telling us to send our children to school, I agree. But the schools should be better. Even though they are free, the children have to pay for uniforms, for books, and we some- times don’t have enough to eat, so to get money for those things, we have to go to our relatives and get goats to sell.” —28-year-old Turkana woman (KII 2015) 242 Appendixes “They keep on telling us ‘send your children to school, educate them’. … Where is the money to send the children to school? They say, you don’t have to pay school fees. But what about exam fees, uniforms, books, all that costs money. So does the transport to school.… I have tried but always we run out of money and our children are chased from school.” —76-year-old Turkana male (KII 2015) The Turkana are ambivalent toward education. On one hand, education is increasingly seen as one of the only strategies available to get out of poverty. On the other, it is seen as something imposed upon them by local and national leaders who are opposed to the pastoralist lifestyle that characterizes Turkana communities. Still, the fact that families are ready to spend a significant part of their household income on education is evi- dence that it increasingly represents a viable investment for Turkana households. These investments in education are linked to an acknowledgment among the Turkana that in their current situation, they are only able to access menial and unskilled jobs, even in their own county. Some respondents made this observation in anger; others simply acknowledged that the low level of education among the Turkana overall limited their ability to access jobs requiring skilled labor. It is a positive development that the Turkana community increasingly views educa- tion as a viable strategy, but if they do not receive any benefit from it, an important opportunity will have been missed. Several respondents noted that the quality of edu- cation is poor and that teachers regularly beat the children. Furthermore, corruption and nepotism often play roles in the attribution of skilled or salaried jobs, increasing frustration among the local community and decreasing incentives for them to investing their meager capital into education. Support to Extended Family Another important expense for Turkana households is the support they give to extended groups of relatives for general expenses like education or in the event of particular dis- tress, such as drought. These expenses often prevent households from saving money, and yet this kind of support given to extended family can be considered an insurance strategy because it is reciprocal among members of the awi who have left to “the city” as well as the rest of the awi who take care of the livestock. An individual with a job is not only expected to share with their immediate family—their wife, children, brothers, parents, and sisters—but also with distant relatives who might ask for money to pay tuition or uniforms, which makes saving difficult. I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 243 This relationship is reciprocal in the sense that the Turkana can turn to their pastoralist awi if an unexpected event such as an illness or death puts them in a difficult financial situation. In such cases, the awi is expected to give them goats in return. “It is up to them. If they refuse, we come back empty-handed. But this does not happen often. Droughts come frequently and, if they refuse too much, so can we. And it does not help the family at all.” —26-year-old Turkana man (KII 2015) Investing in Livestock In line with their pastoralist lifestyle, most Turkana respondents aspire to invest in cattle and livestock should they get enough savings to do so. The Turkana explicitly compare their herds with banks: they consider livestock to be a more reliable savings system than formal banks. The rationale is clear and often well articulated. The Turkana understand that they have the necessary skill set to take care of livestock and to manage their herds. Importantly, it is something that they have control over, unlike bank accounts man- aged by non-Turkana. Even women interviewed for this study who owned small busi- nesses are adamant that if they made enough of a profit from their activities, they would invest in livestock. The first aim of businesses and small income-generating activities is to meet the basic needs of the household and then seek to save enough money to invest in livestock. There is no real business development strategy or investment in business activities because they are seen as a step toward a goal rather than as ends in themselves. Despite frequent droughts, livestock remains the most trusted form of investment among the Turkana, and this limits their ability to invest in other types of productive activities or to further diversify their livelihood strategies. In the current context of repeated drought, their lifestyle may prove difficult to sustain. Due to the lack of infrastructure and investment, the county does not have the capacity to support the traditional livelihoods like agriculture and pastoralism that could respond to the food needs of its growing population (Avery 2014). Yet, these various forms of livelihoods—traditional and more “urban” forms related to the refugee camp—are compatible, just the same way as education and a pastoralist lifestyle are. It is clear from interviews that the Turkana are developing more symbiotic relations between these two sets of livelihoods. While their lifestyle is still characterized by survival and resilience, they have developed strategies of coexistence and balance in order to benefit from both lifestyles at the level of the awi. 244 Appendixes Social Relations: Positive Interactions and Forms of Violence Emergence of New Social Interactions Around the Camp Qualitative interviews demonstrate that relationships between the refugee and Turkana communities are complex. On one hand, many describe an array of “positive” relation- ships between these communities, including friendships, acts of hospitality, and inter- marriages. These connections can provide mutual support in situations of distress. The lack of resentment toward refugees among the host community is striking. Many say that the refugees should be welcomed and hosted. The presence of the camp, a multieth- nic and multilingual space located in a remote area, means that the Turkana have been exposed to an unprecedented level of multiculturalism, which translates into various aspects of their daily lives. Exposure to multiple languages is one obvious consequence, and the Turkana who work in the camp have learned multiple languages in order to be able to communicate with the refugees living near them, particularly the Sudanese. One teacher notes that her Turkana students are now multilingual, speaking Somali, Dinka, and Amhara. Various forms of sociability have taken root, linking the communities together. Several respondents note that they had developed friendships at the camp. This was especially true for the women who spend a lot of time there and who regularly interact with the same families. Young people are also influenced by this proximity to various cultures. As the same teacher notes: “The life of Turkana has changed a lot, for good and for bad. Turkana tend to copy outsiders, their culture, the music, to drink bhang, to smoke sheesha” (KII 2015). Intermarriage is an important aspect of the social interactions between refugee and Turkana communities. Ohta analyzed this dynamic in 2005 and noted many cases in which Turkana women married or had relationships with refugee men, although the opposite scenario appears to have been much less common. For the most part, inter- marriages do not raise problems, except the payment of the bridewealth, which is neces- sary for the Turkana to recognize a union as official and that sometimes raises tensions between communities (Ohta 2005). Negotiations usually take place in the presence of community leaders who try to mitigate the conflict and have the parties come to an agreement. Ohta’s analysis is confirmed by qualitative research conducted for this study. Turkana respondents mention the fact that Turkana women are increasingly having sexual relationships with or marrying refugees. According to the Turkana, marriage tra- ditions have rapidly evolved due to the contact with the refugee camp. Turkana women sometimes prefer marrying earlier in the camp over waiting for a Turkana man to be able to afford their dowry. I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 245 Violence and Tensions Positive social relationships between communities are built through daily interactions that take place in and around the camp, but violence is also present and visible in a number of forms. Symbolic violence. When the Turkana come to the refugee camp, they are often visibly “inferior”—they beg for work and sometimes for food and support. Many members of the host community are so destitute that they feel dependent on the refugees they host. This symbolic violence is manifest in economic interactions controlled by the refugees who benefit from strong bargaining power against a local community desper- ate for work and income, and it is present in daily interactions between the Turkana and refugee communities: many women describe being insulted and harassed when at the camp. The perception that refugees are wealthier and more powerful than their hosts has bred feelings of frustration, anger, and helplessness among the Turkana, fueled by the fact their communities already feel despised in their own country. “The worst abuses are the instances of matharau, na’imeny. The insults they call out. Most refugees don’t do that, only some of them. But they are the loudest.” —58-year-old Turkana woman (KII 2015) Gender-based and physical violence. Interactions between communities are charac- terized by more direct forms of violence as well, including gender-based violence and open conflict. Women from the Turkana and refugee communities are at risk in and outside the camp. Focus group discussions highlight the fact that sexual harassment is common in the vicinity of the camp. Women are at particular risk when they go to collect firewood along the river or when they stay too late in the camp. It is important to note that women from both refugee and host communities are victims of gender-based violence at the hands of both communities and that the issue crosses the boundaries of the camp. The various accounts by respondents indicate that the brutality of these forms of vio- lence is acute and visible. Claire Mwangi analyzed the forms of gender-based violence in the camp, but only found a very high occurrence of rape. She cites the monthly sta- tistics of the Lutheran World Foundation, which recorded 23 rapes just for the month of July 2012. According to Mwangi, it is likely that these statistics largely underestimate the number of rapes because many women are reluctant to report this type of incident (Mwangi 2012: 27–28.). This phenomenon is particularly worrying in a context where HIV/AIDS is prevalent: rape seriously weakens HIV/AIDS prevention strategies. 246 Appendixes Violent intercommunity confrontations happen on a regular basis, although actors noted that their frequency has decreased since 2010. The most significant and frequent conflict is between the Turkana and some of the South Sudanese refugees, particularly the Dinka. Violence is usually sparked by small-scale incidents at the camp that quickly escalate, potentially resulting in a high casualty rate among communities that are mili- tarized and have access to weapons. Devolution and New Resources: Factors of Change? The recent discovery of natural resources in the county, particularly oil and water, along with the benefits that the county may get from the ongoing devolution, could be seen as promising factors of change for the host and refugee communities. But preliminary assessments show that many obstacles will need to be removed before local communi- ties can truly benefit from these developments. New Resources for the County Water The discovery of natural resources has led to cautious optimism about the development of Turkana County. Underground water was discovered and assessed by RTI Interna- tional and UNESCO in 2013. In his assessment, Avery noted that the RTI/UNESCO’s report on the Lotikipi Basin Aquifer claimed that it “could provide water for Kenya for 70 years” and could open significant prospects for agricultural production in the area (Avery 2014: 1). While boreholes have been built to tap into this resource, the positive impact on local populations is slow in manifesting. Initial optimism has faded since subsequent assessments of the water reserve raised doubts about its potential for human consumption: “The dissolved salt level of this water sample is 10 times the permissible level for rural and community water supply, and 4 times the absolute limit, hence would not be suitable” (Avery 2014: 4–5). Still, in the frame of Kenya’s National Water Master Plan, these resources are good news for the county and will be key for developing irrigated agriculture in Turkana and more broadly, across the county. But without a strong and responsible water manage- ment system, the sustainability of newly discovered water reserves will be endangered (Avery 2014: 3). Furthermore, the type of investments and infrastructure needed to harness the benefits from the water reserves will take several years to be established and bear fruit. Important investments are needed in terms of infrastructure and resource management and planning to avoid future land degradation and the unsustainable use of Turkana’s water. I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 247 Oil Oil was discovered in 2012 in Kenya, and Turkana is the focus of current oil exploration. As noted by Cordaid in their recent assessment of community perceptions about oil in the county: In 2010, Tullow Oil, a British company, acquired rights on the exploration of Turkana, Marsabit, and Baringo Counties. It made its first discovery of oil in Turkana in 2012. Since then, the company has drilled 11 wells in the county, with an estimated 600 million recoverable barrels of crude oil. This discovery accelerated the develop- ment of large infrastructure projects, including, the Lamu Port–South Sudan–Ethiopia Transport Corridor (LAPSSET), which includes significant investments in transport infrastructure. Expectations around the economic consequences of this discovery are extremely high (Cordaid 2015: 10). Cordaid’s assessment precisely reflects perceptions among the Turkana interviewed for this study, highlighting their cautious hope that they will benefit from the discovery and the crucial need for a more constructive and inclusive dialog with local communities already feeling left out (Cordaid 2015: 8). Local populations already expressed their concern that their resources—as well as revenue—would fall in the hands of “down- Kenyans” and foreigners. Distrust for local elites was at play here: respondents believed that the revenue would end up in the hands of local leaders and not be shared with the local community. According to Cordaid, the local community has raised a number of concerns, including: n Lack of access to employment opportunities for the Turkana in this new sector, aside from low-level jobs; n Procurement, tendering, and contracting processes characterized by nepotism and corruption; n Poor or nonexistent community consultation; and n The potentially negative effect of oil exploration on grazing land and land access, especially because the Turkana are known for their collective ownership of land (Cordaid 2015: 8–10). The presence of natural resources does seem to have at least some impact on the capacity of communities to organize and their willingness to raise concerns as an engaged, col- lective group. A focus group discussion respondent notes: “The Turkana people need to learn from the past to not take things lightly. Last year, when they said that the water was not good, Turkana people got angry. They protested. This is what is needed” (KII 2015). 248 Appendixes Devolution and the Benefits for Turkana County Turkana is one of the counties in Kenya that should benefit the most from the current devolution process. Given its size and its low level of development, the county receives the largest share of the overall national budget, second only to Nairobi county, Turkana county received K Sh 10.2 billion (US$102 million) in July 2015 (Samuel Hall 2015: 38). For development expenses, the county budget for 2015/16 will be broken down as illustrated in figure I.2 (Republic of Kenya 2015). While the 2013/14 budget put an emphasis on infrastructure, especially roads (15 per- cent) and water and irrigation (12 percent), the 2015/16 budget shows a clear focus on education, which comprises 19 percent of the development budget. Planning and governance expenses have decreased significantly in terms of share of the budget from 25 and 8 percent in 2013/14 to 11 and 2 percent, respectively, accompanied by a significant increase in the share of public services, decentralized administration, and disaster management from 4 to 19 percent. The county budget is in line with the major needs of Turkana population, especially regarding the delivery of basic public services, such as education (19 percent) and roads and transport (9 percent). However, it is not clear that the allocation of only 3 percent of the budget for the pastoral economy and fisheries really takes into account the particular socioeconomic profile of the Turkana population. Figure I.2. Tukana County Budget Breakdown, 2015/16 Pastoral economy and Governance—2% fisheries—3% Trade, tourism, culture, and Public services, decentralized industrialization—2% administration, and disaster management—19% Land, physical planning and urban areas management—5% Education and social services—19% Health services and sanitation—6% Energy, environment, and natural resources—8% Roads, transport, and public works—9% County assembly—8% Water, irrigation, and agriculture—8% Source: Republic of Kenya 2015.  I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 249 It is noteworthy that the increased budget allocation dedicated to the county and the development opportunities offered by devolution did not raise the hopes or expectations among the Turkana. Most of the respondents do not expect devolution to produce any actual positive impact on their lives due to doubts about effective implementation in the absence of efforts at strengthening of capacities and accountability of local authorities. “Well devolution does not mean that we get better leaders. Yes, the county will have more autonomy but it is the same leaders and now with less account- ability from the center.” —36-year-old member of the host community at Kakuma (KII 2015) Conclusion: The Way Ahead for Kakuma A review of the factors that limit the socioeconomic development of Kakuma points to a number of structural factors related to the degradation of the environment of the Turkana and refugee communities and the history of marginalization and antagonism between the central government and the Turkana. The Turkana’s traditional pastoralist lifestyle is endangered by various dynamics, such as the degradation of grazing land, insecurity, and an increase in population. An analysis of the relationship of the Turkana with the refugees demonstrates that they have developed new livelihood strategies that take into account the presence of the camp and the economic opportunities it offers. Interestingly, at the level of the awi, the Turkana tend to invest in both the pastoralist economy and in more urban lifestyles related to the camp, including access to education for their children. One side of a family supports the other in times of crisis, drought, or distress, and the ability to access either lifestyle is itself an insurance and coping strategy for extended families. n There is no doubt that the establishment of the refugee camp has upset the local balance, but for most part, it has been positive, as Turkana households increasingly reorganize their livelihood strategies around the camp. Social and economic interac- tions between refugees and host communities are complex but mostly cordial—if not friendly. There is, however, a level of symbolic violence emanating from the refugees’ economic domination over the area and the paradoxical situation of the Turkana host community being dependent on the refugees. Physical violence, some- times brutal, also exists in various forms, including gender-based violence within and outside the camp and harassment of Turkana women and children when they are at the camp. 250 Appendixes n Recent developments should lead to significant changes for Turkana County. If man- aged correctly, the exploitation of natural resources—particularly oil and water— could have large-scale economic consequences for the local population in terms of access to employment and new economic activities, including irrigated agriculture. The context of devolution should lead to more substantial investments in public services, addressing one of the key factors currently holding Kakuma back. But the benefits of these promising developments will only reach the Turkana and possibly the refugees if they are properly managed by local authorities who are competent and accountable. At present, expectations among the local population are low. n A discussion of key factors holding Kakuma back highlights some entry points for development interventions that would benefit host and refugee communities: n There is a need for local and central authorities to go beyond their general opposi- tion to the Turkana’s pastoralist lifestyle. It is not helpful to simply reject their way of life as backward when, to the contrary, it is actually a key survival strategy. The government’s disregard is evident in the county budget, in which only 3 percent is allocated to the pastoral economy and fisheries. Local populations would benefit from more significant investments in these sectors. n The increasing investment in education among the Turkana is an important oppor- tunity that should be exploited. If the community does not see any benefit from their investments in education, economic priorities will dominate. A particular focus should be put on the quality of education, the motivation of teachers, and pedagogical skills. n Gender-based violence is a worrying trend that cuts across communities. It bears significant risks for the psychological and physical health of Turkana and refugee women alike. A lack of police protection and incentives to report gender-based vio- lent crimes is particularly concerning. Mechanisms to report gender-based violence that are currently in place at the camp could be made available to Turkana women. Prevention campaigns and an increase in the police force should be considered for dangerous areas. n The community’s engagement around questions of local governance, including devolution and natural resources management, should be significantly enhanced. Mechanisms are needed to ensure the meaningful participation of the Turkana com- munity regarding oil exploration and extraction as well as projects related to water management. The lack of skills and education among the Turkana could be over- come through on-the-job training schemes focused on the specific skills needed for both ventures. I. Background Paper: Socioeconomic Mapping and Analysis of Kakuma 251 References Avery, S. 2012. Lake Turkana and the Lower Omo: Hydrological Impacts of Major Dam and Irrigation Development. African Studies Centre, University of Oxford. http://www.africanstudies.ox.ac.uk/sites/ sias/files/documents/Volume%20I%20Report.pdf. ———. 2013. What Future for Turkana? The Impact of Hydropower and Irrigation Development on the World’s Largest Desert Lake. African Studies Centre. Oxford: University of Oxford. http://www.afri- canstudies.ox.ac.uk/sites/sias/files/documents/WhatFutureLakeTurkana-%20update.pdf. ———. 2014. “The Turkana Aquifer Discoveries and Development Proposals.” Drylands Learning and Capacity Building Initiative for Improved Policy and Practice in the Horn of Africa Discussion Brief. http://www.fao.org/fileadmin/user_upload/drought/docs/Final_discussion%20paper_Turkana%20 aquifer.pdf. Betts, A., L. Bloom, J. Kaplan, and Naohiko Omata. 2014. Refugees Economies: Rethinking Popular Assump- tions. Humanitarian Innovation Project, Refugee Studies Centre, Oxford Department of International Development. Oxford: University of Oxford. https://www.rsc.ox.ac.uk/files/publications/other/refu- gee-economies-2014.pdf. Cordaid. 2015. Oil Exploration in Kenya: Success Requires Consultation. Assessment of Community Per- ceptions of Oil Exploration in Turkana County, Kenya. Cordaid. https://www.cordaid.org/media/ medialibrary/2015/09/Turkana_Baseline_Report_DEF-LR_Cordaid.pdf. Crisp, J. 2002. “No Solution in Sight: The Problem of Protracted Refugee Situations in Africa.” CCIS Working Paper 68, Center for Comparative Immigration Studies, University of California San Diego. IFAD (Industrial Fund for Agricultural Development). 2000. The Rangelands of Arid and Semi-Arid Areas. Rome, Italy: IFAD. Hendrickson, D., R. Mearns, and J. Armon. 1996. “Livestock Raiding Among the Pastoral Turkana of Kenya.” IDS Bulletin 27(3). ———. 1998. “The Changing Nature of Conflict and Famine Vulnerability: The Case of Livestock Raiding in Turkana District, Kenya.” Disasters 22(3): 185–99. HRW (Human Rights Watch). 2015. There Is No Time Left: Climate Change, Environmental Threats and Human Rights in Turkana County. Human Rights Watch. KNBS (Kenya National Bureau of Statistics). 2013. Exploring Kenya’s Inequality—Pulling Apart or Pooling Together? Nairobi, Kenya: KNBS. Mwangi, Claire Waithira. 2012. Women Refugees and Sexual Violence in Kakuma Camp, Kenya: Invisible Rights, Justice, Protracted Protection and Human Insecurity. Erasmus University. 252 Appendixes Ohta, I. 2005. “Multiple Socio-Economic Relationships Improvised Between the Turkana and Refugees in Kakuma Area, Northwestern Kenya. ” In Displacement Risks in Africa, edited by I. Ohta and Y. Gebre, 315–36. Kyoto University Press and Transpacific Press. Oka, R. 2011. “Unlikely Cities In The Desert: The Informal Economy As Causal Agent For Permanent ‘Urban’ Sustainability In Kakuma Refugee Camp, Kenya.” Urban Anthropology and Studies of Cultural Systems and World Economic Development 40(3–4): 223–62. Pérouse de Montclos, M., and P. W. Kagwanja. 2000. “Refugee Camps or Cities? Camps in Northern Kenya. ” Journal of Refugee Studies 13(2): 212–13. Pragya. n.d. “Conflict Assessment—Northern Kenya.” Pragya. https://www.pragya.org/doc/Conflict_ Assessment_Report.pdf. Reid, Robin S., and James E. Ellis. 1995. “Impacts of Pastoralists on Woodlands in South Turkana, Kenya: Livestock—Mediated Tree Recruitment.” Ecological Applications 5: 978–92. Republic of Kenya. 2015. Turkana County Government: 2015–2016 Budget Estimates. http://mipakani.net/ sites/default/files/uploaded_pdf_documents/Turkana%20County%20Budget%20Estimates%202015- 2016%2C%2027-6-2015%20-%20FINAL.pdf. Samuel Hall. 2014. Devolution and Refugee Management in Kenya: A Work in Progress. Part B. Report prepared for Danish Refugee Council, ReDSS. ———. 2015. Devolution in Kenya: Opportunity for Transitional Solutions for Refugees? Report prepared for Danish Refugee Council, ReDSS. https://drc.dk/media/1419712/final_devolution_report_230715. pdf. Watson, D., and J. van Binsbergen. 2008. Livestock Market Access and Opportunities in Turkana, Kenya. ILRI Research Report 3, Nairobi, Kenya: International Livestock Research Institute. World Bank. 2015. “Early Childhood Development.” World Bank, Washington, DC. http://web.world- bank.org/WBSITE/EXTERNAL/TOPICS/EXTCY/EXTECD/0,,contentMDK:20207804~menuPK:52 8430~pagePK:148956~piPK:216618~theSitePK:344939,00.html. WFP (World Food Program) and UNHCR (United Nations High Commissioner for Refugees). 2014. “UNHCR/WFP Joint Assessment Mission: Dadaab and Kakuma Refugee Camps.” October, WFP and UNHCR.